The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data such as feature specifications, ADRs (Architecture Decision Records), and requirement documents to generate implementation plans.
Ingestion points: The skill reads external project documentation during the 'Readiness Check' and 'Independent Design Review' phases as described in SKILL.md and design-reviewer-prompt.md.
Boundary markers: There are no explicit delimiters or 'ignore' instructions defined to prevent instructions embedded within external specifications from influencing the agent's behavior during the planning phase.
Capability inventory: The skill's primary capability is writing implementation plans to the docs/plans/ directory. It does not perform autonomous code execution or network operations.
Sanitization: The skill does not perform explicit sanitization of the input text from external documents before including it in the generated plan, though it emphasizes 'exact copy' only for specific pinned values.
Context: Given the skill's purpose as a planning tool that hands off output for human review, the risk associated with processing external requirements is negligible.

writing-plans