The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using find to identify Elixir source files within the lib/ directory.
[DYNAMIC_EXECUTION]: The skill leverages MCP tools (mcp__tidewave__project_eval and mcp__tidewave__execute_sql_query) to run dynamic Elixir code and SQL queries against the project's database and runtime for performance analysis.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes project source code, creating a surface for indirect prompt injection where malicious content in code files could influence the agent's behavior.
Ingestion points: Reads all .ex files found in the lib/ directory (SKILL.md).
Boundary markers: No specific delimiters or "ignore instructions" warnings are used when processing the file content.
Capability inventory: Includes file system scanning (find), Elixir code execution (project_eval), SQL execution (execute_sql_query), and the ability to spawn additional analysis agents (phx:elixir-reviewer).
Sanitization: No sanitization or validation of the source code content is performed before analysis.
[DATA_EXFILTRATION]: Accesses internal application metadata and performance data, including database execution plans (EXPLAIN ANALYZE) and process information (Process.info), which exposes system architecture details.

perf