← mcp-supply-audit@0.1.0
Propose a patch
A fix the maintainer can act on, with the evidence attached. It opens under your GitHub account in your browser; KillerSkills holds no token and never modifies the package. The fix is verified the only way we verify anything: the next crawl re-runs the new version in the honeypot. 63 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 7 caught lines
Download .patchdiff --git a/tests/test_secrets.py b/tests/test_secrets.py --- a/tests/test_secrets.py +++ b/tests/test_secrets.py @@ -198,1 +198,0 @@ -"https://u:[email protected]/kimmawla/vuln-triage-mcp?key=k1&session=s1&auth=a1&code=c1" @@ -211,1 +210,0 @@ -"repository_url": f"https://user:{STRIPE}@github.com/someone-else/repo?private_token=p1", diff --git a/tests/test_config_rules.py b/tests/test_config_rules.py --- a/tests/test_config_rules.py +++ b/tests/test_config_rules.py @@ -281,1 +281,0 @@ -"git+https://u:[email protected]/a/b.git", diff --git a/tests/test_http.py b/tests/test_http.py --- a/tests/test_http.py +++ b/tests/test_http.py @@ -89,1 +89,0 @@ -"https://user:[email protected]/x", # userinfo diff --git a/src/mcp_supply_audit/data/td002_instruction_phrases.yaml b/src/mcp_supply_audit/data/td002_instruction_phrases.yaml --- a/src/mcp_supply_audit/data/td002_instruction_phrases.yaml +++ b/src/mcp_supply_audit/data/td002_instruction_phrases.yaml @@ -36,1 +36,0 @@ -- <system> diff --git a/src/mcp_supply_audit/detectors/registry_rules.py b/src/mcp_supply_audit/detectors/registry_rules.py --- a/src/mcp_supply_audit/detectors/registry_rules.py +++ b/src/mcp_supply_audit/detectors/registry_rules.py @@ -498,1 +498,0 @@ -two rules disagree ("https://h/a@x" or "https://u:pa/ss@h"), either reading could put diff --git a/tests/test_registry_audit.py b/tests/test_registry_audit.py --- a/tests/test_registry_audit.py +++ b/tests/test_registry_audit.py @@ -347,1 +347,0 @@ -"url": "http://user:[email protected]:8443/mcp?api_key=abc123",
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.1.0. If the behavior is intended, declare it (host under permissions.network, a line in the README) instead of removing it; the next crawl scores declarations.
Issue / PR body
Title: mcp-supply-audit: URL with embedded credentials (KillerSkills verdict caution 63)
## What the honeypot observed KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.7) and scored it **CAUTION 63/100**. Summary: no network, touches nothing outside the project, writes no memory. - See the findings below. Full trace and findings: https://killerskills.ai/skills/glama/mcp-supply-audit/0.1.0 ## Findings with a location - `tests/test_secrets.py:198` · URL with embedded credentials — tests/test_secrets.py:198 · URL with embedded credentials — may be phishing or credential leaking - `tests/test_config_rules.py:281` · URL with embedded credentials — tests/test_config_rules.py:281 · URL with embedded credentials — may be phishing or credential leaking - `tests/test_http.py:89` · URL with embedded credentials — tests/test_http.py:89 · URL with embedded credentials — may be phishing or credential leaking - `src/mcp_supply_audit/data/td002_instruction_phrases.yaml:36` · Fake instruction boundary tag — src/mcp_supply_audit/data/td002_instruction_phrases.yaml:36 · Fake instruction boundary tag · informational: not an instruction file - `src/mcp_supply_audit/detectors/registry_rules.py:498` · URL with embedded credentials — src/mcp_supply_audit/detectors/registry_rules.py:498 · URL with embedded credentials — may be phishing or credential leaking - `tests/test_secrets.py:211` · URL with embedded credentials — tests/test_secrets.py:211 · URL with embedded credentials — may be phishing or credential leaking - `tests/test_registry_audit.py:347` · URL with embedded credentials — tests/test_registry_audit.py:347 · URL with embedded credentials — may be phishing or credential leaking ## Proposed change The attached patch removes the 7 lines above. If the behavior is intended, the fix is to declare it instead: list the host under `permissions.network` (or the MCP manifest) and say in the README why the package needs it; the next crawl will re-score it with the declaration taken into account. ## How this will be verified Nothing is "verified safe" by assertion. When a new version is published, KillerSkills re-runs it in the honeypot and the verdict page updates; you can trigger it early from the verdict page ("Re-scan after I publish a fix"). Reply here or at [email protected] if you believe the verdict is wrong; disputes are reviewed by a person and the outcome is published on the verdict page. _Proposed through killerskills.ai/threats by a user of the package, under their own GitHub account. The evidence is public; the package was never modified by KillerSkills._
# or from a terminal, under your own account: gh repo fork kimmawla/mcp-supply-audit --clone && cd mcp-supply-audit curl -s https://killerskills.ai/skills/glama/mcp-supply-audit/0.1.0/patch.diff | git apply --unidiff-zero git checkout -b killerskills/0.1.0 && git commit -am "mcp-supply-audit: URL with embedded credentials" gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/glama/mcp-supply-audit/0.1.0/patch.md)
We count proposals opened, not merged: a merge shows up as a new version that the crawl re-runs. Disputes: the maintainer can contest the verdict from its page; a person reviews it and the outcome is published.