The Agent Skills Directory

[SAFE]: The skill facilitates access to djangoproject.com documentation using Unbrowse.ai's infrastructure.
[EXTERNAL_DOWNLOADS]: The skill connects to unbrowse.ai to provide MCP tool functionality. As this is the vendor's own infrastructure (unbrowse-ai), this is a standard and expected operation for this skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes external documentation data from docs.djangoproject.com. While this is a data ingestion surface, it is a read-only operation targeting well-known technical documentation, and the instructions include status checks (succeeded, input_required) to ensure proper handling of results.
[CREDENTIALS_UNSAFE]: The documentation correctly advises users to manage API keys using a secret manager ($UNBROWSE_API_KEY) rather than hardcoding them in configuration files, which is a security best practice.

djangoproject