The Agent Skills Directory

[SAFE]: The skill's primary function is to query official documentation from symone.fr to answer user questions.
[SAFE]: All network operations and tool calls target the official service domain (symone.fr). This includes the use of a JSON-RPC interface via curl for documentation searches, which is used for data retrieval rather than code execution.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external Markdown files and search results.
Ingestion points: Untrusted content is retrieved from https://symone.fr/docs/mcp and various .md files via the fetch_doc tool.
Boundary markers: The instructions include a directive to "Answer briefly... using only what the pages say," which serves as a basic constraint for the agent's response.
Capability inventory: The skill has network access via the defined MCP tools and example curl commands, but does not have file-writing or subprocess execution capabilities beyond these data retrieval operations.
Sanitization: No explicit sanitization or escaping of the retrieved Markdown content is mentioned.

symone-docs-answers