The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted content from the local repository (e.g., project plans, agreements, and the AGENTS.md file) to generate milestones and instructions.
Ingestion points: SKILL.md (Step 1) and references/project-agreements.md ("Gather evidence") direct the agent to inspect architecture boundaries, repository files, and current project state.
Boundary markers: The instructions do not specify the use of delimiters or boundary markers to isolate ingested project data from the agent's core instructions.
Capability inventory: The skill involves reading project files and writing updates to the repository's AGENTS.md file.
Sanitization: The instructions do not mandate specific sanitization or validation of the ingested repository content before it is used to influence planning decisions.

plan-project