github-presence

Pass

Audited by Gen Agent Trust Hub on Nov 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external GitHub surfaces (READMEs, repository metadata, community health files, and templates) to perform its audit and improvement tasks. If these files contain malicious instructions, they could influence the agent's behavior during the audit process.
  • Ingestion points: The skill explicitly gathers the current state of READMEs, metadata, pinned-repo choices, and community files from the target GitHub profile or repository (specified in SKILL.md under 'Inputs to gather').
  • Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions when reading existing repository content.
  • Capability inventory: The skill is capable of generating and modifying files on the filesystem, though it does not appear to have direct shell execution or network capabilities beyond standard platform tools.
  • Sanitization: No specific sanitization or validation of the ingested markdown or metadata content is described to prevent instruction leakage.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 15, 2026, 05:12 PM