brainstorming

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (HIGH): Indirect Prompt Injection vulnerability surface.
  • Ingestion points: The skill instructs the agent to 'Review current project context first (key files/docs, recent commits, existing patterns)' (found in SKILL.md).
  • Boundary markers: No boundary markers or delimiters are specified to isolate data from instructions.
  • Capability inventory: The agent has the authority to write new design documents to the filesystem, perform 'git commit' operations, and create 'git worktree' environments (found in SKILL.md).
  • Sanitization: No sanitization or input validation is implemented for the ingested context.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The README suggests installing the skill from an untrusted source 'https://github.com/okwinds/miscellany' using the 'openskills' utility, which is outside the defined trust scope.
  • [COMMAND_EXECUTION] (LOW): The skill performs Git operations (commit, worktree) which are high-privilege side effects that can be abused if the agent is compromised via prompt injection.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 05:56 AM