The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local Python modules to execute marketing workflows. Evidence includes instructions to run python -m kai.source_data.collect for data gathering and python -m scripts.quality_gates.eco_gate for evaluating work against predefined 'ECO' floors.
[EXTERNAL_DOWNLOADS]: The skill employs a data collector to fetch intelligence from external URLs and third-party platforms such as Yelp, Meta Ads, and SerpApi. This behavior is documented as part of the market research workflow.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from scraped websites and competitor copy, creating a surface for indirect injection. Evidence Chain: 1. Ingestion points: Data collection tools fetch content from external URLs provided at runtime and analyze scraped pages. 2. Boundary markers: Instructions explicitly state that governance documents outrank scraped content and competitor copy. 3. Capability inventory: The skill can execute local scripts and write files to the project workspace. 4. Sanitization: Mitigation relies on instruction prioritization where system rules override external data.

kai