recoup-internal-consulting-content-recycler
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python scripts, such as
pull_posts.pyandpull_engagement.py, found in the project'sintegrations/linkedin/_work/directory to retrieve data. - [INDIRECT_PROMPT_INJECTION]: The skill ingests external content from social media platforms to inform new draft generation, which presents a surface for potential indirect injection.
- Ingestion points: Reads engagement data and candidate post information from
integrations/linkedin/engagement/andcontent/04-published/. - Boundary markers: Present; the instructions explicitly forbid searching other private checkouts and warn against publishing private client work to public repositories.
- Capability inventory: The skill possesses file read/write access to the workspace and executes local Python subprocesses for data retrieval.
- Sanitization: Absent; while content is restructured by sibling skills, there is no specific mention of sanitizing or validating the incoming engagement data strings.
Audit Metadata