codebase-packager

Pass

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local CLI tools including repomix, gitingest, and tldr to perform filesystem traversal, code extraction, and semantic indexing. These tools are used to build context bundles and repository maps.
  • [EXTERNAL_DOWNLOADS]: The skill documents the installation of external dependencies gitingest and llm-tldr via pip, and repomix via npx. The llm-tldr tool also performs a one-time download of a 1.3GB semantic embedding model during initialization.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it aggregates and processes untrusted source code for delivery to an LLM.
  • Ingestion points: Processes entire local or remote repositories through repomix, gitingest, and llm-tldr indexing.
  • Boundary markers: Uses structured XML tags (e.g., <file path="...">) to delimit different files within the generated context bundles.
  • Capability inventory: Performs extensive file reading, metadata extraction, and semantic search over the codebase.
  • Sanitization: Incorporates secretlint to automatically detect and exclude API keys, PII, and other sensitive credentials from the final context bundles.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 24, 2026, 08:36 PM