scan-docs-index

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting content from documentation files and updating a reference document based on that content.\n
  • Ingestion points: Reads markdown files (*.md) from the project root, the docs/ directory, and .claude/docs/ (File: SKILL.md, Phase 1).\n
  • Boundary markers: None identified. The instructions do not define delimiters or specific markers to ignore embedded instructions within the documents being scanned.\n
  • Capability inventory: Uses grep, glob, and ls for filesystem scanning and has file-writing capability to update docs/project-reference/docs-index-reference.md (File: SKILL.md, Phase 4).\n
  • Sanitization: No content sanitization or validation is specified for the text extracted from documents before it is processed or written to the output file.\n- [COMMAND_EXECUTION]: The skill instructs the agent to execute filesystem commands to perform its indexing tasks.\n
  • Evidence: Instructions include using grep, glob, and ls to find patterns, verify counts, and extract cross-references (File: SKILL.md, Phases 1-3).\n
  • Scope: These commands are restricted to scanning project files and documentation, with no evidence of arbitrary code execution or access to system-level directories.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 07:25 AM
Security Audit — agent-trust-hub — scan-docs-index