The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Elixir toolchain commands, specifically mix credo --strict for linting and mix format for code formatting. These are standard development tools and the skill includes instructions to verify tool availability and versions before execution, categorizing missing tools as 'UNVERIFIED' rather than 'PASS'.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local Elixir source files (.ex, .exs) and configuration files (mix.exs).
Ingestion points: Files within the local project workspace (e.g., SKILL.md Step 1).
Boundary markers: The skill uses a structured output template for reporting findings, though it does not explicitly define prompt-level delimiters for the ingested code.
Capability inventory: The agent has the capability to execute shell commands (mix) and modify local files (mix format) as described in SKILL.md and references/tooling/toolchain.md.
Sanitization: The skill relies on the standard Elixir mix tools for parsing and does not mention specific sanitization of file content before analysis.

codeguard-elixir