improving-codebases

Pass

Audited by Gen Agent Trust Hub on Nov 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various system commands through the node:child_process module to interact with Git and run model checkers.
  • Evidence: scripts/checkpoint.mjs and scripts/churn.mjs use spawnSync and execFileSync to run git commands (e.g., git status, git rev-parse, git ls-tree).
  • Evidence: scripts/run-tlc.mjs executes the Choice Architecture (TLA+) Model Checker (tlc) or java via spawn.
  • [DYNAMIC_EXECUTION]: The skill identifies and runs project-specific test commands derived from the repository's configuration files.
  • Evidence: scripts/run-bounded.mjs executes arbitrary commands provided in a request.json file (typically npm test or similar) via spawn. This is regulated by an environment allowlist (ENV_ALLOWLIST) to prevent uncontrolled privilege or credential leakage.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content (source code, comments, Git history) which could contain malicious instructions.
  • Evidence Ingestion: The skill reads file content using git ls-files and readFileSync in SKILL.md (Step 3) and scripts/churn.mjs.
  • Capability Inventory: The skill has the ability to write to the file system and execute shell commands via run-bounded.mjs.
  • Sanitization/Boundaries: The skill includes mandatory security protocols in shared/external-data.md and shared/independent-review.md which explicitly instruct the agent to treat repository content as data and never as instructions. It also uses independent 'goldfish' reviewers (fresh-context subagents) to verify proposed changes, reducing the risk of a single-session injection compromise.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 21, 2026, 01:16 AM