The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides clear instructions for the agent to execute specific shell commands, such as npx cookiy-mcp --client ... -y, to automate the installation and configuration of the Model Context Protocol (MCP) server across different AI environments.
[EXTERNAL_DOWNLOADS]: The skill triggers the download and execution of the cookiy-mcp package from the npm registry and uses Homebrew to install the standalone binary. These resources are authored by the vendor (cookiy-ai) and are required for the skill's primary setup function.
[COMMAND_EXECUTION]: The included installer CLI modifies local application configuration files (such as ~/.claude.json, ~/.cursor/mcp.json, and ~/.cline/mcp_settings.json) to register the Cookiy MCP server. This behavior is documented in the skill's SECURITY.md file and is necessary for cross-client integration.
[INDIRECT_PROMPT_INJECTION]: The synthesize-research-report skill processes untrusted external data in the form of raw interview transcripts and notes.
Ingestion points: External data is ingested from files or provided via $ARGUMENTS during the synthesis workflow in skills/synthesize-research-report/SKILL.md.
Boundary markers: Absent; the instructions do not define specific delimiters to isolate external content from instructions.
Capability inventory: The skill performs file system writes (generating markdown reports) and uses agent orchestration to process data in parallel batches.
Sanitization: Absent; the skill relies on internal LLM guardrails to process qualitative data without explicit filtering or escaping.

cookiy