The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its core workflow of ingesting data from lesson files and writing it to project-wide configuration files.
Ingestion points: Reads data from .plan/local/lessons-learned/ and quality-verification-report.md.
Boundary markers: The instructions lack explicit delimiters or 'ignore instructions' warnings when the agent processes the content of these files.
Capability inventory: The skill has high-privilege write access via the Edit tool to sensitive files such as CLAUDE.md, standards/*.md, and references/*.md across the marketplace and project directory.
Sanitization: The workflow does not specify any sanitization, escaping, or validation steps before data from lessons is promoted to project standards.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute local utility scripts that manage the plan infrastructure.
Evidence: Several commands call python3 .plan/execute-script.py with arguments directed to modules like plan-marshall:manage-references, plan-marshall:manage-lessons, and plan-marshall:manage-logging.

finalize-step-lessons-housekeeping