codely-plan-create-subagents
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a standard utility for creating development plans. It operates on local codebase files and project documentation to provide structured implementation steps.
- [COMMAND_EXECUTION]: The skill recommends using other internal tools, such as the
/codely-git-conventional_commitskill, to automate repository management tasks. These are standard internal skill invocations. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface where it ingests untrusted data from the codebase and subagent reports.
- Ingestion points: Reads project documentation files (
AGENTS.md,docs/*) and processes findings from independent exploration subagents. - Boundary markers: None explicitly defined to separate ingested file content from reasoning instructions.
- Capability inventory: The skill can create new files in
.agents/plans/and trigger other skills for code implementation and git commits. - Sanitization: None detected; the agent is expected to synthesize and summarize findings into a markdown plan.
Audit Metadata