Claude Plugins— plugin

Claude Plugins — independently scanned and version-tracked by SaferSkills.

Is Claude Plugins safe to install?

SaferSkills independently audited Claude Plugins (Plugin) and scored it 15/100 (red). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 3 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

Score
15/100
●●○○○○○○○○
Re-scan~30s
Latest scan
ScannedJun 26, 2026 · 130d ago
Scans run0 over 90 days
Detectors55 checks · 5 categories
Findings3 warnings · 3 high
EngineSaferSkills 2b638c6
View methodology →
SaferSkills installs
This week0
This month0
All time0
CategoryWeightCategory scoreContribution
Securityprompt, exec, net, exfil, eval
35%
0
0.0 pts
Supply chainhash, typosquat, maintainer, lockfile
20%
100
20.0 pts
Maintenancestaleness, pinning, CI
15%
100
15.0 pts
TransparencySKILL.md, perms, README
15%
100
15.0 pts
Communityinstalls, verify, response
15%
100
15.0 pts

Findings & checks · 6 flagged

Securityscore 0 · 6 findings
CRITICALContains a committed GitHub tokenSS-PLUGIN-SECRET-EXFIL-GH-TOKEN-01 · Credential exfiltration · benchmarks/hush/tests/benchmark_evidence.test.js×2
CRITICAL — a committed token is exposed the moment it lands on a public repo — the credential is leaked outright, not merely at risk.
Why it matters

A GitHub token ('ghp_0123456789abcdefghijklmnopqrstuvwxyz',) is committed directly into this plugin's source. Anyone who reads the repo — including everyone who installs the plugin — gets the token, so it must be treated as already compromised.

The exact value spotted
excerptbenchmarks/hush/tests/benchmark_evidence.test.js· javascript
541for (const secret of [
542'sk-ant-api03-AAAABBBBCCCCDDDD',
543'ghp_0123456789abcdefghijklmnopqrstuvwxyz',
544'AKIAIOSFODNN7EXAMPLE',
545'xoxb-1234-5678-abcdefghijklmnop',
Occurrences
2 occurrences · first at L543, also L564
Show all 2 locations
Line
File
L543
benchmarks/hush/tests/benchmark_evidence.test.js
L564
benchmarks/hush/tests/benchmark_evidence.test.js
How to fix
Revoke the leaked token immediately, then remove it from the code and from Git history.
  1. Revoke the token in GitHub settings now — anything pushed is already public.
  2. Delete it from the source and purge it from history; load credentials at runtime from the environment or a secrets store instead.
AvoidGITHUB_TOKEN = "ghp_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
Safer pattern# never commit the token; read it at runtime from the environment GITHUB_TOKEN = os.environ["GITHUB_TOKEN"]
Trace & refs
ruleSS-PLUGIN-SECRET-EXFIL-GH-TOKEN-01sha2566675cd0c365d2c7d…rubric 365aacaView on GitHub
HIGHReads your SSH private keySS-PLUGIN-SECRET-EXFIL-SSH-01 · Credential exfiltration · benchmarks/hush/tests/benchmark_evidence.test.js
HIGH — SSH keys are high-value but their blast radius depends on what they authorize, so this is high rather than critical.
Why it matters

This plugin references an SSH private-key path or a private-key file header (assert.ok(!clean('key -----BEGIN RSA PRIVATE KEY…). An SSH private key authenticates you to servers and Git remotes, so code that reads it can impersonate you wherever that key is trusted.

The exact value spotted
excerptbenchmarks/hush/tests/benchmark_evidence.test.js· javascript
549assert.ok(!clean(`token: ${secret}`).includes(secret.slice(-12)), `${secret} survived`);
550}
551assert.ok(!clean('key -----BEGIN RSA PRIVATE KEY-----\nMIIabc\n-----END RSA PRIVATE KEY-----
… (23 chars elided on L551)
552assert.ok(clean('mail me at [email protected]').includes('<email>'));
553});
Occurrences
1 occurrence · at L551
How to fix
Remove the code that reads the private key; delegate authentication to the SSH agent or the system git client.
  1. Delete any direct read of id_rsa / id_ed25519 or other key files.
  2. Authenticate through the SSH agent or `git` so the private key never enters plugin memory or an outbound request.
Avoidkey = open(os.path.expanduser("~/.ssh/id_rsa")).read() requests.post(url, data={"key": key})
Safer pattern# let the SSH agent / git handle auth; never read or send the key subprocess.run(["git", "fetch", remote], check=True)
Trace & refs
ruleSS-PLUGIN-SECRET-EXFIL-SSH-01sha256d4c78a76431cc26e…rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · benchmarks/foreman/pickflow/arms/prose/destination-question.md
MEDIUM — it fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters

The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.

The exact value spotted
excerptbenchmarks/foreman/pickflow/arms/prose/destination-question.md· markdown
13yet; the answer decides how the prompt gets built and delivered, not the
14other way around. When the user already named a destination, use it and skip
15the question. Assemble and deliver only after the choice, and never ask for it
16a second time.
17 
Occurrences
1 occurrence · at L15
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
  1. Delete blanket "don't ask / no need to confirm" directives from the skill.
  2. If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-DONT-ASK-01sha25614b7852251bf4666…rubric 365aacaView on GitHub
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · benchmarks/foreman/pickflow/arms/prose/prepare-increments.md×2
MEDIUM — it fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters

The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.

The exact value spotted
excerptbenchmarks/foreman/pickflow/arms/prose/prepare-increments.md· markdown
6assembly and delivery; it is not a project setting. Ordinary split requests do
7not enable it: never infer review mode from a split alone. A supplied
8destination remains the destination: do not ask again or replace it to
9accommodate review.
10 
Occurrences
2 occurrences · first at L8, also L33
Show all 2 locations
Line
File
L8
benchmarks/foreman/pickflow/arms/prose/prepare-increments.md
L33
benchmarks/foreman/pickflow/arms/prose/prepare-increments.md
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
  1. Delete blanket "don't ask / no need to confirm" directives from the skill.
  2. If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-DONT-ASK-01sha25612211d7a427c799a…rubric 365aacaView on GitHub
Supply chainscore 100 · 0 findings
✓—All supply chain checks passedNo findings in this category for the latest scan.pass
Maintenancescore 100 · 0 findings
✓—All maintenance checks passedNo findings in this category for the latest scan.pass
Transparencyscore 100 · 0 findings
✓—All transparency checks passedNo findings in this category for the latest scan.pass
Communityscore 100 · 0 findings
✓—All community checks passedNo findings in this category for the latest scan.pass
Vendor response · right of reply
Are you the maintainer? Submit a response →

Audit the pieces. Scan the whole. Decide.

~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.