literate-guide

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from codebase files, documentation, and history to generate its narrative output, creating a potential surface for indirect prompt injection if those sources contain malicious instructions.
  • Ingestion points: Codebase files (source code, READMEs, comments), commit messages, and issue trackers as specified in the 'Deep Reading' step.
  • Boundary markers: The instructions lack specific requirements for delimiters or 'ignore' instructions to safeguard the agent when processing external content.
  • Capability inventory: The agent is authorized to read files, execute git log to analyze project history, and write documentation files to the .claude/docs/ directory.
  • Sanitization: No specific sanitization or validation logic is defined for the codebase content before it is incorporated into the narrative.
  • [COMMAND_EXECUTION]: The skill utilizes git log to investigate the history of the codebase and identify major architectural changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 01:25 PM
Security Audit — agent-trust-hub — literate-guide