skills/d3lm/ai-lab/code-review-loop/Gen Agent Trust Hub

code-review-loop

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git diffs and untracked files in the repository under review. This data is interpolated directly into the prompts for both the Codex reviewer and the Claude author models, creating a surface for malicious code to influence model behavior.
  • Ingestion points: Found in resources/review-code.mjs within the snapshot() function which calls git diff, and in resources/prompts/codex-review.md where the {{DIFF}} variable is interpolated.
  • Boundary markers: Content is enclosed within <diff> tags in the prompt templates (e.g., in codex-review.md and claude-address.md).
  • Capability inventory: The agent has access to Bash (for running the check command), Edit, Write, Read, Glob, Grep, and WebFetch tools.
  • Sanitization: No explicit filtering or sanitization of the diff content is performed before it is added to the model context.
  • [COMMAND_EXECUTION]: The skill executes a command specified by the user through the --check argument. This command is executed via sh -c and is also available to the Claude model to run during the review loop. While the skill documentation mentions sandboxing (using sandbox-exec or bwrap), the integrity of this confinement depends on the external agent-orchestrate library.
  • [EXTERNAL_DOWNLOADS]: As described in SKILL.md, the skill may use pnpm dlx to download and execute version-pinned CLI tools if they are not available on the host system, which involves fetching external code from the npm registry at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 07:11 PM