The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses imperative language such as "CRITICAL" and "MUST" to instruct the agent to prioritize fetching external documentation over its internal training data. This is a common design pattern in documentation-focused skills to ensure technical accuracy.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch content from https://www.meilisearch.com/docs/llms.txt. This reference targets the official domain of Meilisearch, a well-known open-source technology service.
[INDIRECT_PROMPT_INJECTION]: The skill incorporates external data into the agent's context by fetching a remote documentation file, which creates a potential surface for indirect prompt injection if the source were compromised.
Ingestion points: The skill triggers a WebFetch call in SKILL.md to retrieve the llms.txt file.
Boundary markers: No specific delimiters or boundary instructions are used to separate the external content from the agent's internal instructions.
Capability inventory: The agent is expected to use the fetched documentation to write code, configure search indices, and manage API keys as described in SKILL.md.
Sanitization: The skill does not implement sanitization or validation for the content returned by the external URL.

meilisearch