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 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

Score
92/100
●●●●●●●●●
↑ +0 since first scan (92 → 92)Re-scan~30s
Latest scan
ScannedJun 24, 2026 · 28d ago
Scans run1 over 90 days
Detectors55 checks · 5 categories
Findings2 warnings · 0 high
EngineSaferSkills 2b638c6
View methodology →
SaferSkills installs
This week0
This month0
All time0
CategoryWeightCategory scoreContribution
Securityprompt, exec, net, exfil, eval
35%
76
26.6 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 · 2 flagged

Securityscore 76 · 2 findings
MEDIUM"Never tell the user" non-disclosure imperative in the skillSS-SKILL-INJECT-IMPERATIVE-01 · Prompt injection · plugins/claude-migrate/docs/EXTENDING.md
MEDIUMit fires on intent; whether the agent honors the non-disclosure imperative depends on the host model.
Why it matters

A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.

The exact value spotted
excerptplugins/claude-migrate/docs/EXTENDING.md· markdown
208- A SOURCE `enumerate` MUST return uuids so `UNNN` (sorted-uuid order) stays stable and dete
… (16 chars elided on L208)
209modes and re-extractions. A connector that returns positional indices breaks determinism.
210- Never leak auth tokens or PII into a contract file, a unit file, a brief, a result artifac
… (16 chars elided on L210)
211Tokens resolve from env or from the pre-authenticated browser at invocation time; PII passes
… (4 chars elided on L211)
212`[REDACTED:*]` set in `references/pii-policy.md`.
Occurrences
1 occurrence · at L210
How to fix
Remove the non-disclosure imperative, or rescope it so it limits output format, not honesty to the user.
  1. Delete any "never reveal / never tell the user" line aimed at the model's own behavior.
  2. Keep legitimate confidentiality rules about external data (e.g. "do not echo API keys"), which are user-protective, not user-deceiving.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-IMPERATIVE-01sha256a2f97976152c0441rubric 365aacaView on GitHub
MEDIUM"Never tell the user" non-disclosure imperative in the skillSS-SKILL-INJECT-IMPERATIVE-01 · Prompt injection · plugins/ultra-analyzer/docs/EXTENDING-SOURCES.md
MEDIUMit fires on intent; whether the agent honors the non-disclosure imperative depends on the host model.
Why it matters

A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.

The exact value spotted
excerptplugins/ultra-analyzer/docs/EXTENDING-SOURCES.md· markdown
111- The universal `skills/connector/SKILL.md` is NEVER modified to support a new source. All s
… (82 chars elided on L111)
112- Never add a source-type enum to `config.yaml` or `state.json` — the connector file is auth
… (10 chars elided on L112)
113- Never leak auth tokens into `connector.md` content, topic files, findings, or the report.
… (48 chars elided on L113)
114- Gate 1 (`/ultra` pre-discover) reviews your `connector.md` for completeness — if it flags
… (51 chars elided on L114)
Occurrences
1 occurrence · at L113
How to fix
Remove the non-disclosure imperative, or rescope it so it limits output format, not honesty to the user.
  1. Delete any "never reveal / never tell the user" line aimed at the model's own behavior.
  2. Keep legitimate confidentiality rules about external data (e.g. "do not echo API keys"), which are user-protective, not user-deceiving.
Framework references
OWASPLLM01ATLASAML.T0051
Trace & refs
ruleSS-SKILL-INJECT-IMPERATIVE-01sha256a2f97976152c0441rubric 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.