sdd-onboard
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The 'Executor Override' section contains instructions that direct the agent to ignore previous constraints ('the gate above does NOT apply to you') and modify its tool-calling behavior based on its perceived role. While functional for internal orchestration, this pattern uses override language typical of injection techniques.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from the user's codebase, which could contain malicious instructions designed to influence agent behavior.
- Ingestion points: The skill scans the local codebase for 'opportunities' in Phase 1 and investigates code in Phase 2.
- Boundary markers: The provided file does not explicitly define delimiters or 'ignore' instructions for the content read from the codebase.
- Capability inventory: The skill has significant filesystem capabilities, including writing artifacts (Phase 3-6), applying code changes (Phase 7), and merging/archiving files (Phase 9).
- Sanitization: There is no explicit evidence of sanitization or validation of the code content before it is processed or used to generate new artifacts.
Audit Metadata