The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple source files, which could contain malicious instructions designed to influence the agent's behavior during the planning process.
Ingestion points: The skill reads from part-08-growth-plan/growth-plan.md, various Part 7 preparation documents, and a brand profile document.
Boundary markers: The instructions do not define clear boundaries or specify that the agent should ignore instructions embedded within the ingested markdown or preparation files.
Capability inventory: The skill utilizes Bash, Write, and Edit tools, and invokes local scripts such as pdf-generator.py and engagement-state.py to produce deliverables and update state.
Sanitization: There is no evidence of input validation or sanitization of the content extracted from the strategy documents before it is passed to downstream tools or scripts.

yearly-planner