codemap

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled helper script, codemap.mjs, using the node runtime. This script performs local file system operations such as directory traversal, file hashing, and writing state files (.slim/codemap.json) and documentation templates (codemap.md). These operations are restricted to the repository root provided as an argument.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it ingests untrusted data from the repository structure and generates documentation that influences future agent context.\n
  • Ingestion points: Repository file names, directory structure, and file content (processed by sub-agents) as referenced in SKILL.md and scripts/codemap.mjs.\n
  • Boundary markers: None explicitly defined in the instruction files to isolate ingested data from instructions.\n
  • Capability inventory: The skill can write to the filesystem (codemap.md, AGENTS.md, .slim/codemap.json) and instruct the agent to spawn sub-tasks based on discovered content.\n
  • Sanitization: No explicit sanitization or escaping of file names or extracted content is performed before interpolation into documentation files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 12:41 PM