skills/dafang/codestable/cs-req/Gen Agent Trust Hub

cs-req

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python utility script (.codestable/tools/search-yaml.py) to search through project metadata and utilizes git log to track recent implementation changes. These operations are performed on local files to maintain documentation accuracy and are typical for development tools.
  • [PROMPT_INJECTION]: The skill processes user-provided materials, such as product ideas and feedback, which serves as a surface for indirect prompt injection.
  • Ingestion points: Phase 2 reading of 'user materials' including oral descriptions, product ideas, and user feedback.
  • Boundary markers: None identified in the provided instructions.
  • Capability inventory: The skill can write to local requirement files and execute local scripts or Git commands.
  • Sanitization: No explicit sanitization or validation of the ingested user content is specified before it is processed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 12:39 PM
Security Audit — agent-trust-hub — cs-req