Caveat verdict

evalscope

skill-evalscope

88
🟢 Trusted
No high-risk patterns surfaced by the deep scan — automated capability review, not behavioral proof.

Receives external input AND uses eval

LLM evaluation CLI using the published evalscope package; runs benchmarks against user-configured endpoints with user own keys.

Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

45
security
30
transparency
70
maintenance

What it does

These are capability combinations: each listed behavior occurs in the skill, but Caveat detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.

Capability combination critical

Receives external input AND uses eval — the remote code-injection pattern (data-flow not verified)

LLM01 · LLM05 · ASI01 · ASI05

Permission integrity

Installs packages at runtime — transitive dependencies are not auditable

package_install

Findings (1)

Pattern match medium

HTTP request to bare IP address — common in malicious payloads

perf-reference.md · prose · downgraded · http://127.0.0.1

Permissions & capabilities

No declared permissions — minimal attack surface.

package_installnetwork_indynamic_evalcredential_access
Check another skill Browse the registry Auditing your own skills or configs? Use the API