The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The documentation references installation of the unifapi-agent/agents package via npx and connection to the mcp.unifapi.com server. These resources are provided by the skill's author to facilitate data retrieval.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection due to its core function of processing untrusted external data.
Ingestion points: Data is retrieved from external providers via the maps/search, local/search, and seo/serp operations in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the fetched search results.
Capability inventory: The skill is restricted to read-only tools and possesses no file-write, subprocess, or dynamic code execution capabilities across its scripts.
Sanitization: No explicit sanitization or filtering of retrieved data is described.
[SAFE]: The skill implements robust security guardrails, emphasizing a 'read-only' posture and prohibiting platform manipulation. It also gathers context from local project files like .agents/product-marketing.md to ensure the audit is grounded in the user's intended source of truth.

listing-accuracy-audit