skills/jorgemenadev/skills/andy-post/Gen Agent Trust Hub

andy-post

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external "raw ideas" from users to generate content for social media posts.
  • Ingestion points: SKILL.md (Step 2) takes user-provided ideas for conversion into distinct drafts.
  • Boundary markers: The skill implements a mandatory human-in-the-loop step in SKILL.md (Step 3), which halts automated execution until an operator explicitly selects a draft.
  • Capability inventory: The skill has the ability to create, schedule, and publish content via manage_post and upload binary media via manage_media.
  • Sanitization: No explicit input sanitization or delimiter-based isolation for the user-supplied text is mentioned.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the service provider's official API endpoint for post management and media handling.
  • Endpoint: https://app.andypartner.com/api/mcp as documented in SKILL.md and references/lane.md.
  • This communication is the primary function of the skill and targets a well-known service domain.
  • [COMMAND_EXECUTION]: The skill leverages specific MCP tools to perform its tasks.
  • Tools: get_context, list_connections, list_posts, manage_post, and manage_media are used to interface with the workspace and social connections.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 12:27 AM
Security Audit — agent-trust-hub — andy-post