The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/review_plan.py utilizes subprocess.run() to execute an external CLI tool named codex. This execution is used to perform automated quality reviews of planning documents.
[EXTERNAL_DOWNLOADS]: The skill requires an external dependency, the codex CLI tool, to be available on the host system to perform high-reasoning tasks during the 'Quality Assessment' mode.
[DATA_EXFILTRATION]: During the review process, the scripts/review_plan.py script sends the contents of project planning files (including API contracts and business logic) to an external service via the codex CLI. This is an intended functional behavior for the skill's assessment mode.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes untrusted data from generated plan files.
Ingestion points: Reads content from plan.md, shared.md, backend.md, frontend.md, flows.md, and schema.mmd in the project directories.
Boundary markers: No explicit delimiters or boundary markers are used when interpolating file content into the review prompts.
Capability inventory: The orchestrator can launch parallel sub-agents with varying capabilities and the skill can execute shell commands via local Python scripts.
Sanitization: No input sanitization or validation is performed on the content of the markdown files before they are processed by the orchestrator or passed to the reviewer.

domain-planner