Vulnerability-Lookup

CVE-2026-62674 (GCVE-0-2026-62674)

Vulnerability from cvelistv5 – Published: 2026-08-21 17:47 – Updated: 2026-08-25 17:43

Title

Omnigent: Shared Agent Bundle Overwrite Leads to Authenticated Runner RCE

Summary

Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, PUT /sessions/{session_id}/agent checks LEVEL_EDIT permission for a session but does not reject a bound shared or template agent whose agent.session_id is None. An authenticated user with edit access to a session can replace that shared agent bundle through omnigent/server/routes/sessions.py, add a stdio MCP server, and cause later sessions that use the shared agent to launch an attacker-controlled command through omnigent/tools/mcp.py. The command executes with the Omnigent runner process permissions and can expose files, credentials, workspace data, internal services, and runner availability. This issue is fixed in version 0.3.0.

Severity

9 (Critical)


                        
                          CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H

SSVC

Exploitation: none Automatable: no Technical Impact: total

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2026-08-25 17:38 UTC

CWE

CWE-94 - Improper Control of Generation of Code ('Code Injection')

Assigner

GitHub_M mitre root

References

4 references

URL	Tags
https://github.com/omnigent-ai/omnigent/security/…	x_refsource_CONFIRM
https://github.com/omnigent-ai/omnigent/pull/1418	x_refsource_MISC
https://github.com/omnigent-ai/omnigent/commit/25…	x_refsource_MISC
https://github.com/omnigent-ai/omnigent/releases/…	x_refsource_MISC

Impacted products

1 product

Vendor	Product	Version	CPE status
omnigent-ai	omnigent	Affected: < 0.3.0	guessed

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-62674",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-08-25T17:38:24.682130Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-08-25T17:43:46.509Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "omnigent",
          "vendor": "omnigent-ai",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 0.3.0"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, PUT /sessions/{session_id}/agent checks LEVEL_EDIT permission for a session but does not reject a bound shared or template agent whose agent.session_id is None. An authenticated user with edit access to a session can replace that shared agent bundle through omnigent/server/routes/sessions.py, add a stdio MCP server, and cause later sessions that use the shared agent to launch an attacker-controlled command through omnigent/tools/mcp.py. The command executes with the Omnigent runner process permissions and can expose files, credentials, workspace data, internal services, and runner availability. This issue is fixed in version 0.3.0."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 9,
            "baseSeverity": "CRITICAL",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "LOW",
            "scope": "CHANGED",
            "userInteraction": "REQUIRED",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-94",
              "description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-08-21T17:47:37.976Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h"
        },
        {
          "name": "https://github.com/omnigent-ai/omnigent/pull/1418",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/omnigent-ai/omnigent/pull/1418"
        },
        {
          "name": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b"
        },
        {
          "name": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0"
        }
      ],
      "source": {
        "advisory": "GHSA-jrrm-9hc7-2v3h",
        "discovery": "UNKNOWN"
      },
      "title": "Omnigent: Shared Agent Bundle Overwrite Leads to Authenticated Runner RCE"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-62674",
    "datePublished": "2026-08-21T17:47:37.976Z",
    "dateReserved": "2026-07-14T20:22:04.395Z",
    "dateUpdated": "2026-08-25T17:43:46.509Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "epss": {
      "cve": "CVE-2026-62674",
      "date": "2026-09-17",
      "epss": "0.00343",
      "percentile": "0.27763"
    },
    "nvd": {
      "cve": {
        "affected": [
          {
            "affectedData": [
              {
                "product": "omnigent",
                "vendor": "omnigent-ai",
                "versions": [
                  {
                    "status": "affected",
                    "version": "\u003c 0.3.0"
                  }
                ]
              }
            ],
            "source": "security-advisories@github.com"
          }
        ],
        "cveTags": [],
        "descriptions": [
          {
            "lang": "en",
            "value": "Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, PUT /sessions/{session_id}/agent checks LEVEL_EDIT permission for a session but does not reject a bound shared or template agent whose agent.session_id is None. An authenticated user with edit access to a session can replace that shared agent bundle through omnigent/server/routes/sessions.py, add a stdio MCP server, and cause later sessions that use the shared agent to launch an attacker-controlled command through omnigent/tools/mcp.py. The command executes with the Omnigent runner process permissions and can expose files, credentials, workspace data, internal services, and runner availability. This issue is fixed in version 0.3.0."
          }
        ],
        "id": "CVE-2026-62674",
        "lastModified": "2026-08-25T18:17:58.760",
        "metrics": {
          "cvssMetricV31": [
            {
              "cvssData": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "HIGH",
                "baseScore": 9.0,
                "baseSeverity": "CRITICAL",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "LOW",
                "scope": "CHANGED",
                "userInteraction": "REQUIRED",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H",
                "version": "3.1"
              },
              "exploitabilityScore": 2.3,
              "impactScore": 6.0,
              "source": "security-advisories@github.com",
              "type": "Secondary"
            }
          ],
          "ssvcV203": [
            {
              "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "ssvcData": {
                "id": "CVE-2026-62674",
                "options": [
                  {
                    "exploitation": "none"
                  },
                  {
                    "automatable": "no"
                  },
                  {
                    "technicalImpact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-08-25T17:38:24.682130Z",
                "version": "2.0.3"
              }
            }
          ]
        },
        "published": "2026-08-21T18:16:49.460",
        "references": [
          {
            "source": "security-advisories@github.com",
            "url": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b"
          },
          {
            "source": "security-advisories@github.com",
            "url": "https://github.com/omnigent-ai/omnigent/pull/1418"
          },
          {
            "source": "security-advisories@github.com",
            "url": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0"
          },
          {
            "source": "security-advisories@github.com",
            "url": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h"
          }
        ],
        "sourceIdentifier": "security-advisories@github.com",
        "vulnStatus": "Received",
        "weaknesses": [
          {
            "description": [
              {
                "lang": "en",
                "value": "CWE-94"
              }
            ],
            "source": "security-advisories@github.com",
            "type": "Secondary"
          }
        ]
      }
    },
    "vulnrichment": {
      "containers": {
        "adp": [
          {
            "metrics": [
              {
                "other": {
                  "content": {
                    "id": "CVE-2026-62674",
                    "options": [
                      {
                        "Exploitation": "none"
                      },
                      {
                        "Automatable": "no"
                      },
                      {
                        "Technical Impact": "total"
                      }
                    ],
                    "role": "CISA Coordinator",
                    "timestamp": "2026-08-25T17:38:24.682130Z",
                    "version": "2.0.3"
                  },
                  "type": "ssvc"
                }
              }
            ],
            "providerMetadata": {
              "dateUpdated": "2026-08-25T17:43:07.399Z",
              "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
              "shortName": "CISA-ADP"
            },
            "title": "CISA ADP Vulnrichment"
          }
        ],
        "cna": {
          "affected": [
            {
              "product": "omnigent",
              "vendor": "omnigent-ai",
              "versions": [
                {
                  "status": "affected",
                  "version": "\u003c 0.3.0"
                }
              ]
            }
          ],
          "descriptions": [
            {
              "lang": "en",
              "value": "Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, PUT /sessions/{session_id}/agent checks LEVEL_EDIT permission for a session but does not reject a bound shared or template agent whose agent.session_id is None. An authenticated user with edit access to a session can replace that shared agent bundle through omnigent/server/routes/sessions.py, add a stdio MCP server, and cause later sessions that use the shared agent to launch an attacker-controlled command through omnigent/tools/mcp.py. The command executes with the Omnigent runner process permissions and can expose files, credentials, workspace data, internal services, and runner availability. This issue is fixed in version 0.3.0."
            }
          ],
          "metrics": [
            {
              "cvssV3_1": {
                "attackComplexity": "LOW",
                "attackVector": "NETWORK",
                "availabilityImpact": "HIGH",
                "baseScore": 9,
                "baseSeverity": "CRITICAL",
                "confidentialityImpact": "HIGH",
                "integrityImpact": "HIGH",
                "privilegesRequired": "LOW",
                "scope": "CHANGED",
                "userInteraction": "REQUIRED",
                "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H",
                "version": "3.1"
              }
            }
          ],
          "problemTypes": [
            {
              "descriptions": [
                {
                  "cweId": "CWE-94",
                  "description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
                  "lang": "en",
                  "type": "CWE"
                }
              ]
            }
          ],
          "providerMetadata": {
            "dateUpdated": "2026-08-21T17:47:37.976Z",
            "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
            "shortName": "GitHub_M"
          },
          "references": [
            {
              "name": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h",
              "tags": [
                "x_refsource_CONFIRM"
              ],
              "url": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h"
            },
            {
              "name": "https://github.com/omnigent-ai/omnigent/pull/1418",
              "tags": [
                "x_refsource_MISC"
              ],
              "url": "https://github.com/omnigent-ai/omnigent/pull/1418"
            },
            {
              "name": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b",
              "tags": [
                "x_refsource_MISC"
              ],
              "url": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b"
            },
            {
              "name": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0",
              "tags": [
                "x_refsource_MISC"
              ],
              "url": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0"
            }
          ],
          "source": {
            "advisory": "GHSA-jrrm-9hc7-2v3h",
            "discovery": "UNKNOWN"
          },
          "title": "Omnigent: Shared Agent Bundle Overwrite Leads to Authenticated Runner RCE"
        }
      },
      "cveMetadata": {
        "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "assignerShortName": "GitHub_M",
        "cveId": "CVE-2026-62674",
        "datePublished": "2026-08-21T17:47:37.976Z",
        "dateReserved": "2026-07-14T20:22:04.395Z",
        "dateUpdated": "2026-08-25T17:43:46.509Z",
        "state": "PUBLISHED"
      },
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}

FKIE_CVE-2026-62674

Vulnerability from fkie_nvd - Published: 2026-08-21 18:16 - Updated: 2026-08-25 18:17

Severity

9.0 (Critical) - CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H

Summary

References

	URL	Tags
	security-advisories@github.com	https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b
	security-advisories@github.com	https://github.com/omnigent-ai/omnigent/pull/1418
	security-advisories@github.com	https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0
	security-advisories@github.com	https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h

Impacted products

	Vendor	Product	Version

JSON

To clipboard

{
  "affected": [
    {
      "affectedData": [
        {
          "product": "omnigent",
          "vendor": "omnigent-ai",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 0.3.0"
            }
          ]
        }
      ],
      "source": "security-advisories@github.com"
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, PUT /sessions/{session_id}/agent checks LEVEL_EDIT permission for a session but does not reject a bound shared or template agent whose agent.session_id is None. An authenticated user with edit access to a session can replace that shared agent bundle through omnigent/server/routes/sessions.py, add a stdio MCP server, and cause later sessions that use the shared agent to launch an attacker-controlled command through omnigent/tools/mcp.py. The command executes with the Omnigent runner process permissions and can expose files, credentials, workspace data, internal services, and runner availability. This issue is fixed in version 0.3.0."
    }
  ],
  "id": "CVE-2026-62674",
  "lastModified": "2026-08-25T18:17:58.760",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "HIGH",
          "baseScore": 9.0,
          "baseSeverity": "CRITICAL",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "HIGH",
          "privilegesRequired": "LOW",
          "scope": "CHANGED",
          "userInteraction": "REQUIRED",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H",
          "version": "3.1"
        },
        "exploitabilityScore": 2.3,
        "impactScore": 6.0,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      }
    ],
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2026-62674",
          "options": [
            {
              "exploitation": "none"
            },
            {
              "automatable": "no"
            },
            {
              "technicalImpact": "total"
            }
          ],
          "role": "CISA Coordinator",
          "timestamp": "2026-08-25T17:38:24.682130Z",
          "version": "2.0.3"
        }
      }
    ]
  },
  "published": "2026-08-21T18:16:49.460",
  "references": [
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/omnigent-ai/omnigent/pull/1418"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Received",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-94"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Secondary"
    }
  ]
}

GHSA-JRRM-9HC7-2V3H

Vulnerability from github – Published: 2026-09-02 21:55 – Updated: 2026-09-02 21:55

Summary

Omnigent: Shared Agent Bundle Overwrite Leads to Authenticated Runner RCE

Details

Summary

An authenticated user with edit access to their own session can overwrite a shared/template agent by uploading a full agent bundle through PUT /sessions/{session_id}/agent.

Shared/template agents are shown as not MCP-editable, but this upload path still accepts a replacement bundle. By adding a stdio MCP server to the shared agent, the attacker can cause future runner sessions using that shared agent to start an attacker-controlled command.

Details

The vulnerable endpoint is the full agent bundle upload route: PUT /sessions/{session_id}/agent This route checks whether the caller can edit the session: LEVEL_EDIT session permission check But it does not check whether the bound agent is a shared/template agent. Shared/template agents have:

agent.session_id is None

The API already exposes that these agents are not meant to be MCP-editable: mcp_servers_editable is false for shared/template agents The direct MCP edit endpoint correctly blocks shared/template agents: session_mcp_servers.py shared-agent guard

if agent.session_id is None:
    raise OmnigentError("Built-in agents are read-only through this endpoint.")

The full bundle upload endpoint is missing that same guard. It stores the uploaded bundle and updates the agent row: agent_store.update(agent.id, new_loc) For shared/template agents, this changes the shared agent itself, not only the attacker's session. This becomes command execution because stdio MCP servers start the configured command as a runner subprocess: tools/mcp.py stdio subprocess launch

Suggested fix: add the same agent.session_id is None guard to PUT /sessions/{session_id}/agent.

Impact

This is authenticated RCE against Omnigent runner hosts. A normal authenticated user with edit access to their own session can poison a shared/template agent bundle, even though that agent is shown as not MCP-editable. The poisoned bundle can then be used by future sessions created from the same shared agent.

Because stdio MCP servers run as local subprocesses, the attacker-controlled command executes with the permissions of the Omnigent runner process. In a shared or company-hosted deployment, this can let an attacker read runner-accessible files and credentials, modify workspace data, disrupt runner availability, pivot to internal services reachable from the runner, or establish an interactive remote shell if outbound networking is allowed.

This is high impact because it turns normal shared-agent session access into command execution on runner infrastructure used by other sessions.

Severity

9.0 (Critical)


                  
                    CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "omnigent"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.3.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-62674"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-94"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-09-02T21:55:11Z",
    "nvd_published_at": "2026-08-21T18:16:49Z",
    "severity": "CRITICAL"
  },
  "details": "### Summary\n\nAn authenticated user with edit access to their own session can overwrite a shared/template agent by uploading a full agent bundle through `PUT /sessions/{session_id}/agent`.\n\nShared/template agents are shown as not MCP-editable, but this upload path still accepts a replacement bundle. By adding a `stdio` MCP server to the shared agent, the attacker can cause future runner sessions using that shared agent to start an attacker-controlled command.\n\n### Details\n\nThe vulnerable endpoint is the full agent bundle upload route: [`PUT /sessions/{session_id}/agent`](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L18979-L19077) \nThis route checks whether the caller can edit the session: [`LEVEL_EDIT` session permission check](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L19003-L19025)\nBut it does not check whether the bound agent is a shared/template agent.\nShared/template agents have:\n```python\nagent.session_id is None\n```\nThe API already exposes that these agents are not meant to be MCP-editable: [`mcp_servers_editable` is false for shared/template agents](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L18842-L18857)\nThe direct MCP edit endpoint correctly blocks shared/template agents: [`session_mcp_servers.py` shared-agent guard](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/session_mcp_servers.py#L170-L183)\n```python\nif agent.session_id is None:\n    raise OmnigentError(\"Built-in agents are read-only through this endpoint.\")\n```\nThe full bundle upload endpoint is missing that same guard. It stores the uploaded bundle and updates the agent row: [`agent_store.update(agent.id, new_loc)`](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L19050-L19077)\nFor shared/template agents, this changes the shared agent itself, not only the attacker\u0027s session.\nThis becomes command execution because `stdio` MCP servers start the configured command as a runner subprocess: [`tools/mcp.py` stdio subprocess launch](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/tools/mcp.py#L971-L1021)\n\nSuggested fix: add the same `agent.session_id is None` guard to `PUT /sessions/{session_id}/agent`.\n\n### Impact\n\nThis is authenticated RCE against Omnigent runner hosts. A normal authenticated user with edit access to their own session can poison a shared/template agent bundle, even though that agent is shown as not MCP-editable. The poisoned bundle can then be used by future sessions created from the same shared agent.\n\nBecause `stdio` MCP servers run as local subprocesses, the attacker-controlled command executes with the permissions of the Omnigent runner process. In a shared or company-hosted deployment, this can let an attacker read runner-accessible files and credentials, modify workspace data, disrupt runner availability, pivot to internal services reachable from the runner, or establish an interactive remote shell if outbound networking is allowed.\n\nThis is high impact because it turns normal shared-agent session access into command execution on runner infrastructure used by other sessions.",
  "id": "GHSA-jrrm-9hc7-2v3h",
  "modified": "2026-09-02T21:55:11Z",
  "published": "2026-09-02T21:55:11Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-62674"
    },
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/pull/1418"
    },
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/omnigent-ai/omnigent"
    },
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Omnigent: Shared Agent Bundle Overwrite Leads to Authenticated Runner RCE"
}

PYSEC-2026-3874

Vulnerability from pysec - Published: 2026-09-10 09:44 - Updated: 2026-09-10 11:02

Details

Summary

An authenticated user with edit access to their own session can overwrite a shared/template agent by uploading a full agent bundle through PUT /sessions/{session_id}/agent.

Details

agent.session_id is None

if agent.session_id is None:
    raise OmnigentError("Built-in agents are read-only through this endpoint.")

Suggested fix: add the same agent.session_id is None guard to PUT /sessions/{session_id}/agent.

Impact

This is high impact because it turns normal shared-agent session access into command execution on runner infrastructure used by other sessions.

Severity

9.0 (Critical)


                  
                    CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H

Impacted products

Name	purl
omnigent	pkg:pypi/omnigent

Aliases

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "omnigent",
        "purl": "pkg:pypi/omnigent"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.3.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ],
      "versions": [
        "0.0.1rc1",
        "0.0.1rc2",
        "0.1.0",
        "0.1.0rc1",
        "0.1.0rc2",
        "0.1.0rc3",
        "0.1.0rc4",
        "0.1.1",
        "0.1.1rc2",
        "0.2.0",
        "0.2.0rc1",
        "0.3.0rc1"
      ]
    }
  ],
  "aliases": [
    "CVE-2026-62674",
    "GHSA-jrrm-9hc7-2v3h"
  ],
  "details": "### Summary\n\nAn authenticated user with edit access to their own session can overwrite a shared/template agent by uploading a full agent bundle through `PUT /sessions/{session_id}/agent`.\n\nShared/template agents are shown as not MCP-editable, but this upload path still accepts a replacement bundle. By adding a `stdio` MCP server to the shared agent, the attacker can cause future runner sessions using that shared agent to start an attacker-controlled command.\n\n### Details\n\nThe vulnerable endpoint is the full agent bundle upload route: [`PUT /sessions/{session_id}/agent`](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L18979-L19077) \nThis route checks whether the caller can edit the session: [`LEVEL_EDIT` session permission check](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L19003-L19025)\nBut it does not check whether the bound agent is a shared/template agent.\nShared/template agents have:\n```python\nagent.session_id is None\n```\nThe API already exposes that these agents are not meant to be MCP-editable: [`mcp_servers_editable` is false for shared/template agents](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L18842-L18857)\nThe direct MCP edit endpoint correctly blocks shared/template agents: [`session_mcp_servers.py` shared-agent guard](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/session_mcp_servers.py#L170-L183)\n```python\nif agent.session_id is None:\n    raise OmnigentError(\"Built-in agents are read-only through this endpoint.\")\n```\nThe full bundle upload endpoint is missing that same guard. It stores the uploaded bundle and updates the agent row: [`agent_store.update(agent.id, new_loc)`](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/server/routes/sessions.py#L19050-L19077)\nFor shared/template agents, this changes the shared agent itself, not only the attacker\u0027s session.\nThis becomes command execution because `stdio` MCP servers start the configured command as a runner subprocess: [`tools/mcp.py` stdio subprocess launch](https://github.com/omnigent-ai/omnigent/blob/10f5ae3110e162f96fb99bb6662c581abdc56cd6/omnigent/tools/mcp.py#L971-L1021)\n\nSuggested fix: add the same `agent.session_id is None` guard to `PUT /sessions/{session_id}/agent`.\n\n### Impact\n\nThis is authenticated RCE against Omnigent runner hosts. A normal authenticated user with edit access to their own session can poison a shared/template agent bundle, even though that agent is shown as not MCP-editable. The poisoned bundle can then be used by future sessions created from the same shared agent.\n\nBecause `stdio` MCP servers run as local subprocesses, the attacker-controlled command executes with the permissions of the Omnigent runner process. In a shared or company-hosted deployment, this can let an attacker read runner-accessible files and credentials, modify workspace data, disrupt runner availability, pivot to internal services reachable from the runner, or establish an interactive remote shell if outbound networking is allowed.\n\nThis is high impact because it turns normal shared-agent session access into command execution on runner infrastructure used by other sessions.",
  "id": "PYSEC-2026-3874",
  "modified": "2026-09-10T11:02:18.247657Z",
  "published": "2026-09-10T09:44:59.639583Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/security/advisories/GHSA-jrrm-9hc7-2v3h"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-62674"
    },
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/pull/1418"
    },
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/commit/25a22dc9e6da4648d23749f0a589e47e6aed991b"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/omnigent-ai/omnigent"
    },
    {
      "type": "WEB",
      "url": "https://github.com/omnigent-ai/omnigent/releases/tag/v0.3.0"
    },
    {
      "type": "PACKAGE",
      "url": "https://pypi.org/project/omnigent"
    },
    {
      "type": "ADVISORY",
      "url": "https://github.com/advisories/GHSA-jrrm-9hc7-2v3h"
    }
  ],
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Omnigent: Shared Agent Bundle Overwrite Leads to Authenticated Runner RCE"
}

Sightings

Author	Source	Type	Date	Other

Nomenclature

Seen: The vulnerability was mentioned, discussed, or observed by the user.
Confirmed: The vulnerability has been validated from an analyst's perspective.
Published Proof of Concept: A public proof of concept is available for this vulnerability.
Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
Not confirmed: The user expressed doubt about the validity of the vulnerability.
Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.

Detection rules are retrieved from Rulezet.

The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base model, served locally by ML-Gateway. They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.

Related by attack behaviour

Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.

Action not permitted

CVE-2026-62674 (GCVE-0-2026-62674)

FKIE_CVE-2026-62674

GHSA-JRRM-9HC7-2V3H

Summary

Details

Impact

PYSEC-2026-3874

Summary

Details

Impact

Tags

Sightings

Nomenclature

Related by attack behaviour