The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the execution of the @claude-flow/cli package via npx from the NPM registry to manage memory namespaces. This is a standard operation for Node.js-based management tools.
[COMMAND_EXECUTION]: The skill uses the Bash tool to run the harness command-line utility with various subcommands (oia-manifest, threat-model, mcp-scan) to analyze infrastructure and threat surfaces.
[PROMPT_INJECTION]: The skill provides an interface for auditing local project files, which constitutes an indirect prompt injection surface.
Ingestion points: The skill processes project files at the provided <path> using the harness tool suite.
Boundary markers: No specific delimiters or instructions to ignore embedded content are described in the skill definition.
Capability inventory: The skill possesses shell command execution capabilities via the Bash tool to perform its auditing tasks.
Sanitization: No explicit input sanitization or content filtering of the scanned files is documented, which is standard for static analysis tools.

harness-oia-audit