GHSA-HXH5-GP44-RC6P
Vulnerability from github – Published: 2026-11-19 16:03 – Updated: 2026-11-19 16:03What's wrong
verify_bundle documents that it verifies "a bundle's DSSE envelope AND each per-claim signature", lists "per-claim signature failure" among its raised errors, and promises that partial verification never surfaces "an authenticated bundle that contains some invalid claims" as success. The code does not do this. The JSON-LD predicate built by JSONLDExporter._claim_node carries no signature material (there is no signatureBundle field on the node), so the verify loop has nothing cryptographic to check per claim. It only re-derives each claim's canonical-statement digest from the node fields and compares it to the signed subject list:
chain_input = _signing.canonical_statement({...node fields...}, node.get("evidence") or {})
expected = hashlib.sha256(chain_input).hexdigest()
if subjects[subject_name] != expected:
raise BundleVerificationError(...)
That digest check is self-referential: the exporter signs both the subject digests and the predicate nodes with the same key. It confirms the bundle is internally consistent as built, not that each claim's purported asserter ever validly signed it. The existing test test_tampered_claim_text_in_predicate_fails asserts a "digest mismatch", confirming digest-only verification is the actual contract.
Severity and impact
Medium. All trust in a bundle collapses to the single exporter key. Whoever holds it (or a malicious exporter) can emit a bundle containing claims whose asserter signatures are missing, invalid, or forged, and verify_bundle returns success. A relying party reading the docstring believes per-claim authenticity was verified when it was not. Security-sensitive because this is the trust boundary the SCITT-style export exists to provide.
Location
mareforma/export_bundle.py:20module docstring claims verification checks "every per-claim signature inside predicate"mareforma/export_bundle.py:183verify_bundledocstring promises per-claim signature verification and a per-claim signature failure errormareforma/export_bundle.py:254verify loop only recomputes and compares canonical-statement digestsmareforma/exporters/jsonld.py:178_claim_nodeemits no signature field, so the predicate carries nothing to verify per claim
Fix direction
Two honest options. Either export each claim's signature bundle in the JSON-LD predicate and have verify_bundle cryptographically verify it (including the claim_id and subject binding) against the claim's asserter or validator key, raising on the first failure as documented. Or, if per-claim verification is out of scope for the bundle format, correct the module docstring and the verify_bundle docstring to state exactly what is verified: the bundle-level DSSE signature plus internal subject-digest consistency, with per-claim asserter signatures explicitly not checked.
Do red/green TDD: add a regression test that fails on main, then make it pass.
Quality bar
SOLID, DRY, KISS. Clean, readable, well-tested code. No new errors, warnings, or test failures; the full suite stays green on all supported Python versions.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "mareforma"
},
"ranges": [
{
"events": [
{
"introduced": "0.3.0"
},
{
"fixed": "0.3.8"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-71536"
],
"database_specific": {
"cwe_ids": [
"CWE-347"
],
"github_reviewed": true,
"github_reviewed_at": "2026-11-19T16:03:39Z",
"nvd_published_at": "2026-11-18T23:17:21Z",
"severity": "HIGH"
},
"details": "## What\u0027s wrong\n\n`verify_bundle` documents that it verifies \"a bundle\u0027s DSSE envelope AND each per-claim signature\", lists \"per-claim signature failure\" among its raised errors, and promises that partial verification never surfaces \"an authenticated bundle that contains some invalid claims\" as success. The code does not do this. The JSON-LD predicate built by `JSONLDExporter._claim_node` carries no signature material (there is no `signatureBundle` field on the node), so the verify loop has nothing cryptographic to check per claim. It only re-derives each claim\u0027s canonical-statement digest from the node fields and compares it to the signed subject list:\n\n```python\nchain_input = _signing.canonical_statement({...node fields...}, node.get(\"evidence\") or {})\nexpected = hashlib.sha256(chain_input).hexdigest()\nif subjects[subject_name] != expected:\n raise BundleVerificationError(...)\n```\n\nThat digest check is self-referential: the exporter signs both the subject digests and the predicate nodes with the same key. It confirms the bundle is internally consistent as built, not that each claim\u0027s purported asserter ever validly signed it. The existing test `test_tampered_claim_text_in_predicate_fails` asserts a \"digest mismatch\", confirming digest-only verification is the actual contract.\n\n## Severity and impact\n\n**Medium.** All trust in a bundle collapses to the single exporter key. Whoever holds it (or a malicious exporter) can emit a bundle containing claims whose asserter signatures are missing, invalid, or forged, and `verify_bundle` returns success. A relying party reading the docstring believes per-claim authenticity was verified when it was not. Security-sensitive because this is the trust boundary the SCITT-style export exists to provide.\n\n## Location\n\n- `mareforma/export_bundle.py:20` module docstring claims verification checks \"every per-claim signature inside predicate\"\n- `mareforma/export_bundle.py:183` `verify_bundle` docstring promises per-claim signature verification and a per-claim signature failure error\n- `mareforma/export_bundle.py:254` verify loop only recomputes and compares canonical-statement digests\n- `mareforma/exporters/jsonld.py:178` `_claim_node` emits no signature field, so the predicate carries nothing to verify per claim\n\n## Fix direction\n\nTwo honest options. Either export each claim\u0027s signature bundle in the JSON-LD predicate and have `verify_bundle` cryptographically verify it (including the claim_id and subject binding) against the claim\u0027s asserter or validator key, raising on the first failure as documented. Or, if per-claim verification is out of scope for the bundle format, correct the module docstring and the `verify_bundle` docstring to state exactly what is verified: the bundle-level DSSE signature plus internal subject-digest consistency, with per-claim asserter signatures explicitly not checked.\n\nDo red/green TDD: add a regression test that fails on `main`, then make it pass.\n\n## Quality bar\n\nSOLID, DRY, KISS. Clean, readable, well-tested code. No new errors, warnings, or test failures; the full suite stays green on all supported Python versions.",
"id": "GHSA-hxh5-gp44-rc6p",
"modified": "2026-11-19T16:03:39Z",
"published": "2026-11-19T16:03:39Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/mareforma/mareforma/security/advisories/GHSA-hxh5-gp44-rc6p"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-71536"
},
{
"type": "WEB",
"url": "https://github.com/mareforma/mareforma/pull/58"
},
{
"type": "WEB",
"url": "https://github.com/mareforma/mareforma/commit/20caa4a5a9557a140a09a5d2c1c4a6a2124c0cc3"
},
{
"type": "PACKAGE",
"url": "https://github.com/mareforma/mareforma"
},
{
"type": "WEB",
"url": "https://github.com/mareforma/mareforma/releases/tag/v0.3.8"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:N/VI:H/VA:N/SC:N/SI:H/SA:N",
"type": "CVSS_V4"
}
],
"summary": "Mareforma: verify_bundle never checks a per-claim signature"
}
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 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.