krt-docs-chronicler

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads repository content—including source code, configuration files, and commit history—to verify documentation accuracy.
  • Ingestion points: SKILL.md instructs the agent to inspect files, commands, scripts, config, manifests, tests, examples, and recent diffs.
  • Boundary markers: None explicitly defined to separate untrusted repo data from agent instructions.
  • Capability inventory: The skill is capable of modifying existing documentation and creating new documentation files.
  • Sanitization: No explicit sanitization of ingested content is mentioned, though the skill contains instructions to avoid inventing commands or architecture.
  • [EXTERNAL_DOWNLOADS]: The skill references several external documentation standards and frameworks, including Google's documentation style guide, Diataxis, and 'Keep a Changelog'. These references are used for conceptual guidance and do not involve the execution of remote code or the download of untrusted scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 08:32 PM
Security Audit — agent-trust-hub — krt-docs-chronicler