The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the youmind CLI and node scripts to interact with the backend service and process data.
Evidence: SKILL.md invokes youmind call for various API operations and node -e to parse message outputs.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @youmind-ai/cli package from the public NPM registry.
Evidence: The workflow in SKILL.md checks for and installs @youmind-ai/cli using npm install -g if it is missing.
[INDIRECT_PROMPT_INJECTION]: The skill processes data returned from the YouMind API, which could theoretically contain instructions intended to influence the agent's behavior.
Ingestion points: The youmind call listMessages command in SKILL.md (Step 5) fetches external content into the agent's context.
Boundary markers: None identified for the raw data ingestion.
Capability inventory: The skill uses youmind CLI commands, npm install, and node -e for execution.
Sanitization: The skill uses a node script to specifically parse and extract tool result blocks from the JSON response, effectively filtering for structured data rather than interpreting general text as instructions.

youmind-slides-generator