The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill's runner script (scripts/run.py) performs network operations to interact with the vendor's API.
Evidence: Sends POST requests to https://ai-skills.ai/api/execute using Python's urllib.request library.
Context: This is the primary function of the skill to process data via the remote service.
[COMMAND_EXECUTION]: The skill documentation instructs users to execute a Python script to initiate the coaching process.
Evidence: python3 scripts/run.py --params '{}' in SKILL.md.
Context: Standard operation for a developer-oriented skill tool.
[DATA_EXFILTRATION]: The skill transmits potentially sensitive sales data and an API key to the vendor's infrastructure.
Evidence: User-provided inputs from materialText, materialFile, and materialUrl along with the AISKILLS_API_KEY are sent to ai-skills.ai in the request headers and body.
Context: This is the documented and intended behavior of the service; it does not access local system credentials or unauthorized files.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from external sources.
Ingestion points: materialText, materialFile, and materialUrl defined in references/form-schema.json.
Boundary markers: None identified in the provided client-side script.
Capability inventory: Network communication via urllib.request in scripts/run.py.
Sanitization: No sanitization or validation of the input content is performed in the runner script.

ai-sales-coaching