The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the metaharness CLI tool via a subprocess using npx to perform analysis on a specified repository path. This is the primary mechanism for generating the readiness scorecard.
[EXTERNAL_DOWNLOADS]: The skill uses npx, which may fetch the metaharness package from the public npm registry at runtime if it is not available in the local environment.
[DATA_EXPOSURE]: The analysis process involves reading source code and configuration files from the local filesystem at the path provided to the tool to calculate metrics like task coverage and tool safety.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted data from the repository being analyzed.
Ingestion points: Files within the repository path provided as an argument (e.g., source code, scripts, metadata).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the analyzed files are defined in the skill instructions.
Capability inventory: The skill has access to the Bash tool for subprocess execution.
Sanitization: No specific sanitization or validation of the repository content before analysis is described in the provided markdown.

harness-score