The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources such as design documents and wiki pages. While this presents a surface for indirect prompt injection where instructions could be embedded in the design documents, the skill is limited to generating a tasks file and requires explicit user approval before any implementation tasks are executed.
Ingestion points: Reads design.md, AGENTS.md, and project wiki pages.
Boundary markers: The skill does not explicitly define boundary markers for the ingested data.
Capability inventory: Performs file writes to the project documentation directory and calls other agent tools such as implement-task and read-codebase-context.
Sanitization: No specific sanitization of input text is mentioned.
[DATA_EXPOSURE]: The skill reads the codebase context and project-specific documentation to perform its function. This behavior is consistent with its stated purpose of planning features based on the existing environment.

plan-feature