The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Retrieves platform-specific skill documentation from the vendor's GitHub repository to provide tailored advice for specific tools.
[SAFE]: No malicious patterns, such as credential exfiltration, suspicious command execution, or obfuscation, were identified. The skill's operations are consistent with its stated purpose of managing sales data workflows.
[PROMPT_INJECTION]: The skill implements a self-improvement loop using a local reference file. 1. Ingestion points: 'references/learnings.md' is read at the beginning of sessions. 2. Boundary markers: No explicit markers are used to delimit user-contributed findings. 3. Capability inventory: The skill instructions include reading local markdown files and fetching remote content via 'WebFetch'. 4. Sanitization: Appended data is not sanitized, creating a surface for potential indirect prompt injection.

sales-enrich