The Agent Skills Directory

[METADATA_POISONING]: The 'Skill Boundaries' section of the documentation contains misleading information, claiming the skill assists with 'audio production workflows' and 'audio engineering' despite the skill's focus on competitive analysis. This metadata discrepancy could lead to confusion regarding the skill's intended purpose and safety constraints.
[COMMAND_EXECUTION]: The scripts/main.py script includes functionality to read and write JSON files on the local file system based on user-provided file paths. While this is part of the tool's intended use for competitor profiling and matrix generation, it constitutes a file system interaction capability.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input through its documentation prompts and script-based data ingestion. Ingestion points: Data enters the agent's context through industry descriptions provided by the user and JSON files processed by the matrix command in scripts/main.py. Boundary markers: There are no specific delimiters or instructions defined in the documentation or code to protect the agent from instructions embedded within the external industry or competitor data. Capability inventory: The skill's scripts are capable of local file system read/write operations and console output, though they lack network or subprocess execution capabilities. Sanitization: No input validation or sanitization routines are implemented for the data ingested from user prompts or external files.

competitive-analysis