read-docs

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it instructs the agent to process and summarize content from markdown files within a project, which could include untrusted or malicious instructions.
  • Ingestion points: Content is ingested from docs/**/*.md, README.md, CLAUDE.md, and other root-level .md files as defined in the Search Strategy of SKILL.md.
  • Boundary markers: No explicit boundary markers or instructions are provided to the agent to treat documentation content as non-authoritative data or to ignore embedded instructions.
  • Capability inventory: The skill uses the agent's ability to read and summarize files, incorporating external content into its reasoning process.
  • Sanitization: No sanitization, escaping, or validation of file contents is specified.
  • [NO_CODE]: This skill contains no executable code files, scripts, or binaries, and consists solely of markdown instructions, which minimizes the risk of direct malicious code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 09:23 PM