The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill requires a HubSpot Access Token to function. It correctly instructs the user to manage this sensitive credential using a .env file, which is a standard and safe practice for local agent environments to prevent hardcoding secrets.
[COMMAND_EXECUTION]: The skill utilizes uv run to execute its auditing script. This is a legitimate use of a modern Python package manager to ensure a reproducible execution environment with the correct dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill processes data retrieved from the HubSpot API. While this involves ingesting external data, the risk is minimal as the script acts as a read-only analyzer that outputs structured reports. The evidence chain shows clear ingestion points (HubSpot Search API), no high-risk capabilities like dynamic code execution on the ingested data, and specific report formatting logic that isolates external content.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns were detected. The dependencies (requests, python-dotenv) are standard, well-known libraries pinned in the script's metadata.

hubspot-audit