Skills
Navigate
skills/jeremylongshore/tons-of-skills-marketplace/anth-known-pitfalls/Gen Agent Trust Hub

anth-known-pitfalls

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides instructional content for proper API usage and does not contain any executable code or dangerous commands.
  • [EXTERNAL_DOWNLOADS]: The skill references official Anthropic documentation and SDK repositories on GitHub, which are trusted sources for integration guidance.
  • [INDIRECT_PROMPT_INJECTION]: The skill's purpose involves reviewing external code, creating an ingestion surface for untrusted data.
  • Ingestion points: Code and configuration files read during the integration review process (SKILL.md).
  • Boundary markers: The instructions explicitly advise using synthetic fixtures and sandbox environments to isolate the agent from potentially malicious or sensitive data.
  • Capability inventory: The skill is permitted to use Read, Write, Edit, and Grep tools to analyze and modify files.
  • Sanitization: Instructions mandate the use of a redaction policy and emphasize that production content or keys should not be used during diagnostics.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 10:24 AM