beyond-code-plan
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a restricted architectural planning workflow that explicitly forbids the agent from writing or modifying any source code.
- [SAFE]: The skill operates on local project files within a defined directory structure (
.beyond-code/) and does not perform any network operations or access sensitive system credentials. - [INDIRECT_PROMPT_INJECTION]: The skill processes project requirements from
spec.md, which is a potential surface for indirect prompt injection. However, the risk is minimized by the skill's rigid output templates and the requirement for human confirmation before advancing to implementation stages. - Ingestion points:
.beyond-code/<slug>/gate.md,.beyond-code/<slug>/spec.md(Stage 0). - Boundary markers: Absent.
- Capability inventory: File-write operations to
plan.mdandgate.md(Stage 4 and Stage 6). - Sanitization: Absent.
Audit Metadata