constitution

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from existing documentation files and user profiles to generate a project constitution, creating a surface for potential indirect prompt injection.
  • Ingestion points: The skill reads 02-DOCS/wiki/harness/user-profile.md for user context and 02-DOCS/wiki/stack/*.md for existing conventions.
  • Boundary markers: There are no explicit delimiters or specific warnings to ignore embedded instructions when reading these local files.
  • Capability inventory: The skill is restricted to file system operations (reading documentation and writing to 02-DOCS/wiki/sdd/constitution.md and CLAUDE.md). It does not have network access or arbitrary code execution capabilities.
  • Sanitization: No content filtering is performed; however, the skill mandates that the agent show the draft to the user for explicit ratification before it takes effect.
  • [SAFE]: The skill proactively addresses security by establishing a non-negotiable principle that 'No secret is ever committed,' directing secrets to be stored in gitignored environment files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 08:57 AM
Security Audit — agent-trust-hub — constitution