{
  "affected": [
    {
      "affectedData": [
        {
          "product": "PraisonAI",
          "vendor": "MervinPraison",
          "versions": [
            {
              "status": "affected",
              "version": "\u003e= 1.5.1, \u003c 1.7.2"
            }
          ]
        }
      ],
      "source": "security-advisories@github.com"
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "PraisonAI is a multi-agent teams system. From 1.5.1 until 1.7.2, MCPSecurity.evaluatePolicy() in src/praisonai-ts/src/mcp/security.ts invokes the configured credential validator only when AuthMethod is api-key or bearer. Basic and OAuth policies accept any non-empty Authorization header without calling auth.validate(), then return an authenticated result, allowing callers with invalid credentials to access MCP tools and resources protected by those policies. This issue is fixed in version 1.7.2."
    }
  ],
  "id": "CVE-2026-57134",
  "lastModified": "2026-09-15T14:45:28.563",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "NONE",
          "baseScore": 8.2,
          "baseSeverity": "HIGH",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "LOW",
          "privilegesRequired": "NONE",
          "scope": "UNCHANGED",
          "userInteraction": "NONE",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N",
          "version": "3.1"
        },
        "exploitabilityScore": 3.9,
        "impactScore": 4.2,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      }
    ],
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2026-57134",
          "options": [
            {
              "exploitation": "poc"
            },
            {
              "automatable": "yes"
            },
            {
              "technicalImpact": "partial"
            }
          ],
          "role": "CISA Coordinator",
          "timestamp": "2026-09-15T12:36:39.614902Z",
          "version": "2.0.3"
        }
      }
    ]
  },
  "published": "2026-09-15T11:17:10.583",
  "references": [
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/MervinPraison/PraisonAI/commit/709a0387853709a5e2f2c34680fa64cada252c27"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/MervinPraison/PraisonAI/releases/tag/v4.6.60"
    },
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-4qq2-2j2x-x62c"
    },
    {
      "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
      "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-4qq2-2j2x-x62c"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Deferred",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-287"
        },
        {
          "lang": "en",
          "value": "CWE-288"
        },
        {
          "lang": "en",
          "value": "CWE-863"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Secondary"
    }
  ]
}

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 1.7.1"
      },
      "package": {
        "ecosystem": "npm",
        "name": "praisonai"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.5.1"
            },
            {
              "fixed": "1.7.2"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-57134"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-287",
      "CWE-288",
      "CWE-863"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-06-18T14:25:17Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "## Summary\n\nThe published npm package `praisonai` exports an `MCPSecurity` helper described in source as:\n\n```text\nMCP Security - Authentication, authorization, and rate limiting\nProvides security policies for MCP servers.\n```\n\nIts `AuthMethod` type advertises five authentication methods:\n\n```ts\nexport type AuthMethod = \u0027none\u0027 | \u0027api-key\u0027 | \u0027bearer\u0027 | \u0027basic\u0027 | \u0027oauth\u0027;\n```\n\nThe authentication-policy evaluator, however, only validates credentials for `api-key` and `bearer`:\n\n```ts\nif (policy.auth.method === \u0027api-key\u0027 || policy.auth.method === \u0027bearer\u0027) {\n    const valid = policy.auth.validate\n        ? await policy.auth.validate(token)\n        : this.validateApiKey(token);\n\n    if (!valid) {\n        return { allowed: false, reason: \u0027Invalid credentials\u0027 };\n    }\n}\n\nreturn { allowed: true, context: { authenticated: true } };\n```\n\nFor `basic` and `oauth`, any non-empty `Authorization` header skips the supplied `validate` callback and returns allowed. A local PoV configures `auth.validate` to always return `false`; invalid `api-key` and `bearer` credentials are rejected, while invalid `basic` and `oauth` credentials are accepted without calling the validator.\n\nThis is a protection-mechanism failure in the exported npm MCP security helper. It is distinct from the separate issue that the npm `MCPServer` HTTP transport does not enforce authentication by default.\n\n## Technical Details\n\n`SecurityPolicy.auth` accepts both a method and a validator:\n\n```ts\nauth?: { method: AuthMethod; validate?: (token: string) =\u003e Promise\u003cboolean\u003e };\n```\n\n`extractToken()` parses both Bearer and Basic headers:\n\n```ts\nif (auth.startsWith(\u0027Bearer \u0027)) {\n    return auth.slice(7);\n}\nif (auth.startsWith(\u0027Basic \u0027)) {\n    return auth.slice(6);\n}\nreturn auth;\n```\n\nBut `evaluatePolicy()` only calls `policy.auth.validate()` for two methods:\n\n```ts\nif (policy.auth.method === \u0027api-key\u0027 || policy.auth.method === \u0027bearer\u0027) {\n    const valid = policy.auth.validate\n        ? await policy.auth.validate(token)\n        : this.validateApiKey(token);\n\n    if (!valid) {\n        return { allowed: false, reason: \u0027Invalid credentials\u0027 };\n    }\n}\n```\n\nThere is no validation branch for `basic` or `oauth`. After extracting any non-empty token, those methods fall through to the success return:\n\n```ts\nreturn { allowed: true, context: { authenticated: true } };\n```\n\n`check()` then ignores successful authentication context and returns a generic allowed result:\n\n```ts\nreturn { allowed: true, context: { authenticated: false } };\n```\n\nThat context propagation issue is secondary. The security-relevant flaw is that invalid Basic/OAuth credentials are allowed at all.\n\n### Why This Is Not Intended Behavior\n\nThis is not a claim that every `MCPSecurity` user must choose Basic or OAuth. The issue is that the API explicitly exposes those methods as authentication methods and accepts a validator callback for the policy, but the implementation does not call the validator for those methods.\n\nThe control cases prove the intended security behavior:\n\n- Missing Basic credentials are denied as `Authentication required`.\n- Invalid `api-key` credentials are denied as `Invalid credentials`.\n- Invalid `bearer` credentials are denied as `Invalid credentials`.\n\nThe only difference in the vulnerable cases is the selected advertised method. Invalid Basic/OAuth credentials should not become authenticated merely because the method is not listed in the two-method validation branch.\n\nThis also matches MCP authorization guidance. MCP servers acting as resource servers must validate received access tokens; receiving a token is not proof that it is valid or intended for the server.\n\n## PoV\n\nRun from a local reproduction checkout:\n\n```bash\nnode poc/pov_poc.js 1.7.1\n```\n\nThe PoV:\n\n1. Installs `npm:praisonai@1.7.1` into a temporary project with scripts disabled.\n2. Imports `MCPSecurity` from the package root.\n3. Creates one `authenticate` policy per method.\n4. Supplies an `auth.validate` callback that always returns `false`.\n5. Sends invalid `api-key`, `bearer`, `basic`, and `oauth` credentials.\n6. Confirms the missing-header Basic control is still denied.\n\nObserved output summary from `evidence/pov-npm-1.7.1.json`:\n\n```json\n{\n  \"package\": \"praisonai\",\n  \"version\": \"1.7.1\",\n  \"cases\": [\n    {\n      \"method\": \"api-key\",\n      \"validateCalls\": 1,\n      \"allowed\": false,\n      \"reason\": \"Invalid credentials\"\n    },\n    {\n      \"method\": \"bearer\",\n      \"validateCalls\": 1,\n      \"allowed\": false,\n      \"reason\": \"Invalid credentials\"\n    },\n    {\n      \"method\": \"basic\",\n      \"validateCalls\": 0,\n      \"allowed\": true\n    },\n    {\n      \"method\": \"oauth\",\n      \"validateCalls\": 0,\n      \"allowed\": true\n    },\n    {\n      \"method\": \"basic\",\n      \"authorizationHeaderPresent\": false,\n      \"validateCalls\": 0,\n      \"allowed\": false,\n      \"reason\": \"Authentication required\"\n    }\n  ],\n  \"controlsPass\": true,\n  \"vulnerable\": true\n}\n```\n\nThe PoV is local-only. It does not start a server, contact a third-party target, or use live credentials.\n\n## PoC\n\nThe PoV section above contains the local reproduction command, input, and decisive output.\n\n## Impact\n\nA downstream application that uses `MCPSecurity` to protect an HTTP MCP transport, gateway, or equivalent tool/resource endpoint can believe it has enabled Basic or OAuth authentication while accepting any non-empty `Authorization` header.\n\nDepending on the protected MCP tools and resources, this can allow an unauthenticated network caller to:\n\n- list protected tools or resources;\n- call tools that were intended to require authentication;\n- read protected MCP resources;\n- trigger agent/workflow actions exposed behind the security helper; and\n- bypass audit assumptions based on the configured validator.\n\nThis report does not claim that npm PraisonAI wires `MCPSecurity` into the default `MCPServer.startHttp()` path. It is a library-level authentication bypass in an exported security component intended to protect MCP servers.\n\n### Severity\n\nSuggested severity: High.\n\nRationale:\n\n- `AV`: the affected helper is intended to protect MCP server requests and equivalent HTTP security checks.\n- `AC`: a single non-empty Basic or OAuth-style Authorization header is sufficient when such a policy is configured.\n- `PR`: the bypass grants access without valid credentials.\n- `UI`: no maintainer or user interaction is required after deployment.\n- `S`: impact is within the PraisonAI-hosting service and its exposed MCP resources/tools.\n- `C`: protected MCP resources or tool outputs may be disclosed.\n- `I`: protected tool calls may perform state-changing actions depending on the registered tools; the score is conservative because the vulnerable helper is library-level and deployment-dependent.\n- `A`: the PoV does not demonstrate availability impact.\n\nIf a deployment protects high-impact write or execution tools with `MCPSecurity`, maintainers may reasonably score integrity higher.\n\n## Suggested Fix\n\nMake authentication evaluation fail closed for every advertised method.\n\nRecommended:\n\n1. For `authenticate` policies, call `policy.auth.validate(token)` whenever it is provided, regardless of `auth.method`.\n2. If no validator is provided, only fall back to `validateApiKey()` for `api-key` when that behavior is explicitly intended.\n3. For `bearer` and `oauth`, require a validator or a server-side token validation implementation; otherwise deny with a configuration error.\n4. For `basic`, decode the Basic credential safely and pass the decoded username/password or raw credential to a validator; if no validator exists, deny.\n5. Treat unknown or unsupported methods as denied, not allowed.\n6. Return authenticated context from `check()` after a successful authenticate policy instead of replacing it with `{ authenticated: false }`.\n7. Add regression tests proving invalid credentials are rejected for `api-key`, `bearer`, `basic`, and `oauth`, and that each configured validator is called.\n\nMinimal fail-closed shape:\n\n```ts\nif (policy.type === \u0027authenticate\u0027) {\n  if (!policy.auth) return { allowed: false, reason: \u0027Authentication policy is not configured\u0027 };\n\n  const token = request.headers ? this.extractToken(request.headers) : null;\n  if (!token) return { allowed: false, reason: \u0027Authentication required\u0027 };\n\n  if (policy.auth.validate) {\n    const valid = await policy.auth.validate(token);\n    return valid\n      ? { allowed: true, context: { authenticated: true } }\n      : { allowed: false, reason: \u0027Invalid credentials\u0027 };\n  }\n\n  if (policy.auth.method === \u0027api-key\u0027) {\n    return this.validateApiKey(token)\n      ? { allowed: true, context: { authenticated: true } }\n      : { allowed: false, reason: \u0027Invalid credentials\u0027 };\n  }\n\n  return { allowed: false, reason: \u0027Authentication validator required\u0027 };\n}\n```\n\n## Affected Package/Versions\n\n- Repository: `MervinPraison/PraisonAI`\n- Ecosystem: `npm`\n- Package: `praisonai`\n- Component: TypeScript MCP security helper `src/praisonai-ts/src/mcp/security.ts`\n- Published dist path: `node_modules/praisonai/dist/mcp/security.js`\n- Latest npm package validated: `1.7.1`\n- Current `origin/main` validated: `1ad58ca02975ff1398efeda694ea2ab78f20cf3e`\n- `src/praisonai-ts/package.json` at `origin/main`: `praisonai` `1.7.1`\n\nSuggested affected range:\n\n```text\nnpm:praisonai \u003e= 1.5.1, \u003c= 1.7.1\n```\n\nAll published npm `1.x` versions were swept locally:\n\n- `1.0.0` through `1.5.0`: the tested root export was unavailable or `MCPSecurity` was not exported as a constructor.\n- `1.5.1`, `1.5.2`, `1.5.3`, `1.5.4`, `1.6.0`, `1.7.0`, and `1.7.1`: vulnerable.\n\nThe package root re-exports this helper:\n\n```ts\nexport {\n  MCPClient, createMCPClient, getMCPTools,\n  MCPServer, createMCPServer,\n  MCPSession as MCPSessionManager, createMCPSession,\n  MCPSecurity, createMCPSecurity, createApiKeyPolicy, createRateLimitPolicy,\n  type MCPClientConfig, type MCPSession, type MCPTransportType,\n  type MCPServerConfig, type MCPServerTool,\n  type SecurityPolicy, type SecurityResult\n} from \u0027./mcp\u0027;\n```\n\n## Advisory History\n\nVisible PraisonAI advisories and prior submissions were checked. The closest public advisory is `GHSA-98f9-fqg5-hvq5` / `CVE-2026-34953`, but that issue is distinct:\n\n- `GHSA-98f9-fqg5-hvq5` affects the PyPI package and Python `OAuthManager.validate_token()`.\n- This report affects the npm package and TypeScript `src/praisonai-ts/src/mcp/security.ts`.\n- The prior issue accepts arbitrary Bearer tokens because an empty Python token store falls through to `True`.\n- This issue accepts invalid Basic/OAuth credentials because the TypeScript validator callback is never called for those advertised methods.\n- The affected ranges and patched surfaces are different.\n\nThe earlier npm `MCPServer` report is also distinct: it covers missing auth in the HTTP transport by default. This report covers a fail-open branch in the separate exported `MCPSecurity` helper when users attempt to add Basic/OAuth authentication.",
  "id": "GHSA-4qq2-2j2x-x62c",
  "modified": "2026-07-20T21:27:02Z",
  "published": "2026-06-18T14:25:17Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-4qq2-2j2x-x62c"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/MervinPraison/PraisonAI"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "npm PraisonAI MCPSecurity Basic/OAuth authentication policies accept invalid credentials without validation"
}