crabpot-perf-metrics

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill uses established tools like the GitHub CLI to fetch data from a trusted source. No malicious persistence, privilege escalation, or obfuscation was found.
  • [PROMPT_INJECTION]: The skill ingests performance reports from external sources, which creates a potential surface for indirect prompt injection attacks. 1. Ingestion points: reports/crabpot-dashboard-data.json and related JSON profiles in SKILL.md. 2. Boundary markers: The skill does not provide explicit markers or instructions to treat the ingested data as untrusted content. 3. Capability inventory: The skill uses gh api for data retrieval and summarizes output, representing a read-heavy capability tier with network access. 4. Sanitization: No sanitization or validation of the external JSON structure is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 02:50 AM
Security Audit — agent-trust-hub — crabpot-perf-metrics