callidescope-triage

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides documentation for the 'callidescope' tool and does not contain malicious code or dangerous instructions.
  • [EXTERNAL_DOWNLOADS]: The skill references the command npx callidescope --write, which downloads and executes the 'callidescope' package from the public npm registry. This is consistent with the skill's intended purpose as documentation for using the specific tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for interpreting external tool output (e.g., stack depth findings like [DEPTH n > limit]). 1. Ingestion points: Interpretation of console output by the agent as described in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Command execution via npx. 4. Sanitization: Absent. The injection surface is limited to standard developer triage tasks and does not bypass safety guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 01:12 PM
Security Audit — agent-trust-hub — callidescope-triage