groove-admin-help

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Potential for indirect prompt injection via the configuration file. The skill instructions require the agent to read .groove/index.md and interpolate its values into the output. If an attacker can modify this file, they could inject instructions that the agent might follow due to its high capability level.
  • Ingestion points: .groove/index.md (read by the agent to populate the help output).
  • Boundary markers: Absent. There are no delimiters or instructions provided to the agent to treat the configuration values as untrusted data.
  • Capability inventory: The skill requests broad permissions via allowed-tools, including Bash access for git, gh (GitHub CLI), and npx (executing Node packages), along with file system Write and Edit capabilities.
  • Sanitization: Absent. The skill does not perform any validation or escaping of the data read from the configuration file before displaying it.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 08:30 PM