groove-groovebook-publish

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it ingests untrusted data from multiple sources and uses it to drive agent actions without sufficient isolation.
  • Ingestion points: Data enters the agent's context through the $ARGUMENTS variable, the .groove/index.md configuration file, and interactive user prompts via the AskUserQuestion tool.
  • Boundary markers: The skill lacks explicit delimiters or specific 'ignore instructions' markers when processing external data, which could allow malicious instructions in those inputs to influence the agent.
  • Capability inventory: The skill utilizes powerful tools including Bash(git:*), Bash(gh:*), Write, and Edit, which could be exploited to perform unauthorized repository or filesystem operations if an injection occurs.
  • Sanitization: The skill performs basic sanitization for filenames (slugification) and includes a dedicated step requiring the user to manually review and redact project-specific context before the final publication step.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 12:09 AM