The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple local Python scripts located in the integrations/linkedin/_work/ directory to perform LinkedIn publishing, post discovery, engagement monitoring, and lead enrichment.
Evidence: publish.py, pull_posts.py, pull_engagement.py, enrich.py, crossref_attio.py, and make_loop.py.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from an external social media platform (LinkedIn) to monitor engagement and identify lead candidates, creating a surface for indirect prompt injection.
Ingestion points: LinkedIn posts and engagement data (likes, comments, shares) are fetched via pull_posts.py and pull_engagement.py (Step 4 and 5) and processed in SKILL.md.
Boundary markers: Absent. The skill does not define specific delimiters or instructions to ignore embedded commands within the fetched social media content.
Capability inventory: The skill coordinates subprocess calls to Python scripts for publishing and enrichment, and has the ability to write state to distribution.md, linkedin.md, and the email/outbox/ directory.
Sanitization: Absent. There is no mention of sanitization, validation, or filtering for the external content before it is processed by enrichment scripts or used by the agent to determine lead candidates.

recoup-internal-consulting-content-distribution