The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Node.js script included with the skill package to automate the creation of planning artifacts. Evidence: node "<skill-root>/scripts/scaffold-plan.mjs" <slug> [--clear|--unclear] in SKILL.md. This script is used to generate file structures in the .omo/ directory and is a documented part of the skill's utility.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because its primary function is to read and analyze untrusted files from a repository to generate plans. Ingestion points: The agent uses tools like read, grep, glob, and codegraph_explore to ingest codebase content. Capability inventory: The agent has the ability to execute shell commands (via the Node.js script) and spawn sub-agents. Boundary markers: No specific delimiters or instructions to ignore embedded instructions are provided for codebase reading. Sanitization: There is no explicit sanitization or validation of the content read from the codebase before it is processed or used to generate command arguments.

ulw-plan