Vulnerability-Lookup

GHSA-R8J5-8747-88CM

Vulnerability from github – Published: 2026-05-14 20:55 – Updated: 2026-06-09 10:26

Summary

@utcp/http: SSRF via attacker-controlled OpenAPI servers[0].url in HTTP communication protocol

Details

Summary

The @utcp/http package is vulnerable to a blind Server-Side Request Forgery (SSRF) caused by a trust-boundary inconsistency between manual discovery and tool invocation. registerManual() validates the discovery URL against an HTTPS / loopback allowlist, but callTool() reuses the resolved toolCallTemplate.url directly without revalidating, and the OpenApiConverter blindly trusts whatever servers[0].url an attacker-hosted spec declares. An attacker who hosts a malicious OpenAPI spec on a legitimate HTTPS endpoint can declare e.g. servers: [{ url: "http://127.0.0.1:9090" }] or servers: [{ url: "http://169.254.169.254" }]; the converter then produces tools whose URL points at internal services on the agent host.

A separate prefix-bypass also affected the discovery-time check: the previous startsWith('http://localhost') guard let URLs like http://localhost.evil.com through.

Sister advisory

This is the npm/TypeScript counterpart of GHSA-39j6-4867-gg4w / CVE-2026-44661 on the Python utcp-http package. Same vulnerability, same fix shape, same reporter.

Versions and patch state

@utcp/http <= 1.1.1 — vulnerable. Both the loopback-redirect (http://127.0.0.1) and the non-loopback internal-IP variants (e.g. http://169.254.169.254, http://10.0.0.5) succeed. Note: the streamable_http and sse callToolStreaming paths in 1.1.1 are TODO placeholders and don't actually fetch URLs, so the runtime SSRF surface in these protocols is currently confined to discovery — a future implementation must also call ensureSecureUrl before issuing the request.
@utcp/http 1.1.2 — full fix. Runtime revalidation in callTool closes the non-loopback variants; the OpenApiConverter rejects, at conversion time, any spec fetched from a non-loopback source that declares a loopback servers[0].url, closing the loopback-redirect variant.

Impact

A remote attacker who can convince the agent (via the LLM context, prompt injection, or a tool-discovery surface) to register their HTTPS OpenAPI URL can: - Map internal networks behind the agent. - Read AWS/GCP IAM credentials from cloud metadata endpoints (http://169.254.169.254, http://metadata.google.internal). - Reach unauthenticated internal services exposed on loopback (Elasticsearch, Redis HTTP, internal admin panels, the agent's own HTTP server). - Have responses returned to the LLM, which combined with prompt injection enables exfiltration back to the attacker.

Patch

Commit on dev: 21f63e6.

New helper packages/http/src/_security.ts exposes isSecureUrl, isLoopbackUrl, ensureSecureUrl. Hostname-based validation closes the prefix bypass (http://localhost.evil.com → rejected). All three protocols' registerManual now call ensureSecureUrl(url, 'manual discovery'); callTool re-checks the resolved URL with ensureSecureUrl(url, 'tool invocation') immediately before the axios request. OpenApiConverter rejects remote spec → loopback server.

Workarounds

For users who cannot upgrade immediately: - Refuse to call registerManual with any URL controlled by an untrusted party, even over HTTPS. - Restrict outbound network access from the host running the agent so internal addresses (RFC1918, 169.254.0.0/16, loopback) are unreachable.

Credit

Discovered and reported by @YLChen-007 against the Python sibling implementation (universal-tool-calling-protocol/python-utcp#83). The TypeScript port shared the same code shape and the same vulnerability.

Severity

4.7 (Medium)


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

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 1.1.1"
      },
      "package": {
        "ecosystem": "npm",
        "name": "@utcp/http"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.1.2"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-45366"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-05-14T20:55:05Z",
    "nvd_published_at": "2026-05-28T22:17:00Z",
    "severity": "MODERATE"
  },
  "details": "## Summary\n\nThe `@utcp/http` package is vulnerable to a blind Server-Side Request Forgery (SSRF) caused by a trust-boundary inconsistency between manual discovery and tool invocation. `registerManual()` validates the discovery URL against an HTTPS / loopback allowlist, but `callTool()` reuses the resolved `toolCallTemplate.url` directly without revalidating, and the `OpenApiConverter` blindly trusts whatever `servers[0].url` an attacker-hosted spec declares. An attacker who hosts a malicious OpenAPI spec on a legitimate HTTPS endpoint can declare e.g. `servers: [{ url: \"http://127.0.0.1:9090\" }]` or `servers: [{ url: \"http://169.254.169.254\" }]`; the converter then produces tools whose URL points at internal services on the agent host.\n\nA separate prefix-bypass also affected the discovery-time check: the previous `startsWith(\u0027http://localhost\u0027)` guard let URLs like `http://localhost.evil.com` through.\n\n## Sister advisory\n\nThis is the npm/TypeScript counterpart of [GHSA-39j6-4867-gg4w / CVE-2026-44661](https://github.com/universal-tool-calling-protocol/python-utcp/security/advisories/GHSA-39j6-4867-gg4w) on the Python `utcp-http` package. Same vulnerability, same fix shape, same reporter.\n\n## Versions and patch state\n\n- `@utcp/http \u003c= 1.1.1` \u2014 vulnerable. Both the loopback-redirect (`http://127.0.0.1`) and the non-loopback internal-IP variants (e.g. `http://169.254.169.254`, `http://10.0.0.5`) succeed. Note: the `streamable_http` and `sse` `callToolStreaming` paths in 1.1.1 are TODO placeholders and don\u0027t actually fetch URLs, so the runtime SSRF surface in these protocols is currently confined to discovery \u2014 a future implementation must also call `ensureSecureUrl` before issuing the request.\n- `@utcp/http 1.1.2` \u2014 full fix. Runtime revalidation in `callTool` closes the non-loopback variants; the `OpenApiConverter` rejects, at conversion time, any spec fetched from a non-loopback source that declares a loopback `servers[0].url`, closing the loopback-redirect variant.\n\n## Impact\n\nA remote attacker who can convince the agent (via the LLM context, prompt injection, or a tool-discovery surface) to register their HTTPS OpenAPI URL can:\n- Map internal networks behind the agent.\n- Read AWS/GCP IAM credentials from cloud metadata endpoints (`http://169.254.169.254`, `http://metadata.google.internal`).\n- Reach unauthenticated internal services exposed on loopback (Elasticsearch, Redis HTTP, internal admin panels, the agent\u0027s own HTTP server).\n- Have responses returned to the LLM, which combined with prompt injection enables exfiltration back to the attacker.\n\n## Patch\n\nCommit on `dev`: 21f63e6.\n\nNew helper `packages/http/src/_security.ts` exposes `isSecureUrl`, `isLoopbackUrl`, `ensureSecureUrl`. Hostname-based validation closes the prefix bypass (`http://localhost.evil.com` \u2192 rejected). All three protocols\u0027 `registerManual` now call `ensureSecureUrl(url, \u0027manual discovery\u0027)`; `callTool` re-checks the resolved URL with `ensureSecureUrl(url, \u0027tool invocation\u0027)` immediately before the axios request. `OpenApiConverter` rejects remote spec \u2192 loopback server.\n\n## Workarounds\n\nFor users who cannot upgrade immediately:\n- Refuse to call `registerManual` with any URL controlled by an untrusted party, even over HTTPS.\n- Restrict outbound network access from the host running the agent so internal addresses (RFC1918, 169.254.0.0/16, loopback) are unreachable.\n\n## Credit\n\nDiscovered and reported by [@YLChen-007](https://github.com/YLChen-007) against the Python sibling implementation (universal-tool-calling-protocol/python-utcp#83). The TypeScript port shared the same code shape and the same vulnerability.",
  "id": "GHSA-r8j5-8747-88cm",
  "modified": "2026-06-09T10:26:31Z",
  "published": "2026-05-14T20:55:05Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/universal-tool-calling-protocol/typescript-utcp/security/advisories/GHSA-r8j5-8747-88cm"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-45366"
    },
    {
      "type": "WEB",
      "url": "https://github.com/universal-tool-calling-protocol/python-utcp/issues/83"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/universal-tool-calling-protocol/typescript-utcp"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:L/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "@utcp/http: SSRF via attacker-controlled OpenAPI servers[0].url in HTTP communication protocol"
}

CVE-2026-45366 (GCVE-0-2026-45366)

Vulnerability from cvelistv5 – Published: 2026-05-28 20:38 – Updated: 2026-05-29 13:56

Title

typescript-utcp: SSRF via attacker-controlled OpenAPI servers[0].url in HTTP communication protocol

Summary

typescript-utcp is a typescript implementation of UTCP. Prior to 1.1.2, the @utcp/http package is vulnerable to a blind Server-Side Request Forgery (SSRF) caused by a trust-boundary inconsistency between manual discovery and tool invocation. registerManual() validates the discovery URL against an HTTPS / loopback allowlist, but callTool() reuses the resolved toolCallTemplate.url directly without revalidating, and the OpenApiConverter blindly trusts whatever servers[0].url an attacker-hosted spec declares. An attacker who hosts a malicious OpenAPI spec on a legitimate HTTPS endpoint can declare e.g. servers: [{ url: "http://127.0.0.1:9090" }] or servers: [{ url: "http://169.254.169.254" }]; the converter then produces tools whose URL points at internal services on the agent host. This vulnerability is fixed in 1.1.2.

Severity

4.7 (Medium)


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

SSVC

Exploitation: none Automatable: no Technical Impact: partial

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2026-05-29 13:56 UTC

CWE

CWE-918 - Server-Side Request Forgery (SSRF)

Assigner

GitHub_M mitre root Scorecard

References

1 reference

URL	Tags
https://github.com/universal-tool-calling-protoco…	x_refsource_CONFIRM

Impacted products

1 product

Vendor	Product	Version
universal-tool-calling-protocol	typescript-utcp	Affected: < 1.1.2

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-45366",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-05-29T13:56:02.351079Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-05-29T13:56:10.659Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "typescript-utcp",
          "vendor": "universal-tool-calling-protocol",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 1.1.2"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "typescript-utcp is a typescript implementation of UTCP. Prior to 1.1.2, the @utcp/http package is vulnerable to a blind Server-Side Request Forgery (SSRF) caused by a trust-boundary inconsistency between manual discovery and tool invocation. registerManual() validates the discovery URL against an HTTPS / loopback allowlist, but callTool() reuses the resolved toolCallTemplate.url directly without revalidating, and the OpenApiConverter blindly trusts whatever servers[0].url an attacker-hosted spec declares. An attacker who hosts a malicious OpenAPI spec on a legitimate HTTPS endpoint can declare e.g. servers: [{ url: \"http://127.0.0.1:9090\" }] or servers: [{ url: \"http://169.254.169.254\" }]; the converter then produces tools whose URL points at internal services on the agent host. This vulnerability is fixed in 1.1.2."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "HIGH",
            "attackVector": "NETWORK",
            "availabilityImpact": "NONE",
            "baseScore": 4.7,
            "baseSeverity": "MEDIUM",
            "confidentialityImpact": "LOW",
            "integrityImpact": "LOW",
            "privilegesRequired": "NONE",
            "scope": "CHANGED",
            "userInteraction": "REQUIRED",
            "vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:L/A:N",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-918",
              "description": "CWE-918: Server-Side Request Forgery (SSRF)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-05-28T20:38:16.780Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/universal-tool-calling-protocol/typescript-utcp/security/advisories/GHSA-r8j5-8747-88cm",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/universal-tool-calling-protocol/typescript-utcp/security/advisories/GHSA-r8j5-8747-88cm"
        }
      ],
      "source": {
        "advisory": "GHSA-r8j5-8747-88cm",
        "discovery": "UNKNOWN"
      },
      "title": "typescript-utcp: SSRF via attacker-controlled OpenAPI servers[0].url in HTTP communication protocol"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-45366",
    "datePublished": "2026-05-28T20:38:16.780Z",
    "dateReserved": "2026-05-12T00:51:29.085Z",
    "dateUpdated": "2026-05-29T13:56:10.659Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}

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.

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

GHSA-R8J5-8747-88CM

Summary

Sister advisory

Versions and patch state

Impact

Patch

Workarounds

Credit

CVE-2026-45366 (GCVE-0-2026-45366)

Tags

Sightings

Nomenclature

Related by attack behaviour