configure-task-to-pr

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands like rg, find, and git status to perform repository discovery. These commands are used for read-only inspection of the project's metadata, structure, and configuration files to inform the bootstrapping process.
  • [PROMPT_INJECTION]: The skill processes potentially untrusted data from the repository (e.g., README.md, package.json, and contributor guides) to generate localized agent instructions. This represents an indirect prompt injection surface.
  • Ingestion points: Reads high-signal repository files during Phase 1 (Discovery).
  • Boundary markers: No explicit delimiters are used when processing the discovered file content.
  • Capability inventory: Uses bash for metadata discovery and file system writes to create .agents/skills/, .claude/skills/, or .codex/skills/ directories.
  • Sanitization: No specific sanitization or filtering of the ingested file content is performed before generating the new instruction files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 07:39 PM
Security Audit — agent-trust-hub — configure-task-to-pr