openspec-onboard

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes system commands including 'openspec --version' for environment validation and 'git log' to analyze recent repository history.
  • [INDIRECT_PROMPT_INJECTION]: The skill scans for specific comment patterns like TODO, FIXME, and HACK in the user's codebase to suggest tasks, which serves as a potential ingestion point for malicious instructions.
  • Ingestion points: Local source code files containing development comments scanned during task selection (SKILL.md, Phase 2).
  • Boundary markers: Absent; the skill does not define specific delimiters to separate codebase content from agent instructions.
  • Capability inventory: File system read/write access and shell command execution via the openspec and setup-rhdh-skills toolsets (SKILL.md, Phases 4-9).
  • Sanitization: Absent; the skill does not mention validation or escaping of the content retrieved from comments before processing it.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 08:20 PM
Security Audit — agent-trust-hub — openspec-onboard