The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files and configuration, which could be used to influence the agent's reasoning or the content of generated documentation.
Ingestion points: The skill reads several project-specific files including beat/config.yaml, proposal.md, design.md, and other artifacts located in the beat/changes/ directory.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified when the agent processes the content of these external files.
Capability inventory: The skill is authorized to read the local filesystem and write documentation artifacts (such as ADRs and design notes). However, it is explicitly forbidden from writing application code or implementing features.
Sanitization: The instructions do not include steps to sanitize or validate the content of the ingested files (e.g., the rules.adr in beat/config.yaml) before they are applied as constraints to the agent's output.

explore