create-plan

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill is configured to read from ~/.codex/repo_profile.json as a fallback reference. Although this path appears to be a tool-specific configuration directory, accessing files within the user's home directory is a sensitive operation.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted content from the repository to generate its output.
  • Ingestion points: Reads repository documentation (README, ARCHITECTURE, ROADMAP, etc.) and repository-local profile files (REPO_PROFILE.json).
  • Boundary markers: Absent. The skill does not define specific delimiters or instructions to ignore embedded commands within the files it reads.
  • Capability inventory: The skill performs read-only file system access across the repository and specific configuration paths.
  • Sanitization: Absent. There is no explicit instruction to sanitize or validate the content extracted from repository files before using it to construct the final plan.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 09:56 PM
Security Audit — agent-trust-hub — create-plan