The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates a multi-agent workflow where sub-agents independently explore the codebase, creating a surface for instructions embedded in local files to influence agent behavior.
Ingestion points: Sub-agents (feature-solver-1 through 10) utilize Read, Grep, Glob, and LS tools to examine existing codebase files (Step 3).
Boundary markers: The instructions do not define delimiters or "ignore embedded instructions" warnings for sub-agents when they process file content.
Capability inventory: Solver agents possess filesystem discovery and reading capabilities, while the orchestrating agent is granted file creation and modification permissions to execute the synthesized implementation plan (Step 8).
Sanitization: No sanitization, validation, or filtering of the content read from the codebase is performed before it is incorporated into the final implementation plan.

feature-council