skills/liustack/vibemaster/muse/Gen Agent Trust Hub

muse

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses git check-ignore to inspect the local repository state.
  • [COMMAND_EXECUTION]: The skill modifies the .git/info/exclude file to append the .issues/ directory, which updates local git configuration to ignore session drafts.
  • [PROMPT_INJECTION]: (Category 8: Indirect Prompt Injection) The skill is vulnerable to indirect injection because it processes untrusted user prompts and writes the content to the local file system without sanitization.
  • Ingestion points: User-provided creative topics and brainstorming constraints are processed in SKILL.md.
  • Boundary markers: Absent; the content is written directly to Markdown files in the .issues/ directory.
  • Capability inventory: The skill can execute shell commands (git check-ignore), modify project configuration files (.git/info/exclude), and write new files to the workspace (.issues/).
  • Sanitization: No filtering or validation of user input is performed before it is written to the workspace.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 07:14 AM
Security Audit — agent-trust-hub — muse