planning-validate

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. The skill ingests and evaluates data from multiple external project documents, which could be used to deliver hidden instructions to the agent. * Ingestion points: spec.md, plan.md, data-model.md, and the contracts/ directory. * Boundary markers: Absent; the skill does not define clear delimiters or specific instructions to ignore embedded commands within the analyzed artifacts. * Capability inventory: File system read access and output generation based on the analysis of text content. * Sanitization: Absent; there is no mention of content filtering, escaping, or schema validation for the ingested markdown files.
  • [DATA_EXFILTRATION] (SAFE): While the skill reads sensitive planning documents, no patterns of network transmission or data exfiltration were detected in the provided documentation.
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill mentions an external dependency (check-prerequisites.sh) and a repository for context, but these are presented as installation requirements rather than hidden runtime code execution or unauthorized downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:30 PM