The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external project documentation and user-provided artifacts to generate business analysis outputs.
Ingestion points: Reads files from docs/specs/{module}/*.md, {idea-file}, and {pbi-file}.
Boundary markers: The instructions do not define clear delimiters or warnings for the agent to ignore instructions that might be embedded within these documentation files.
Capability inventory: The skill can perform file system operations (read/write), execute Glob searches, and is authorized to use spawn_agent to create subagents.
Sanitization: There is no evidence of input validation or sanitization for the content ingested from the project documents.
[PROMPT_INJECTION]: The skill contains meta-instructions that attempt to override standard behavior, such as 'Ignore Claude-specific mode-switch instructions when they appear.' While intended for cross-platform compatibility, this matches the pattern of behavior-override instructions.
[COMMAND_EXECUTION]: The skill utilizes Glob patterns for dynamic file discovery in the docs/ directory and explicitly authorizes the use of the spawn_agent tool to delegate tasks to subagents.

business-analyst