The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to clone code from a personal GitHub repository (https://github.com/brandonwise/humanizer.git). This source is not a verified organization or well-known service.
[COMMAND_EXECUTION]: The installation instructions include running npm install and npm install -g . on the cloned source code. This process can automatically execute arbitrary shell scripts defined in the repository's package.json (such as preinstall or postinstall hooks).
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze external text files (node src/cli.js analyze -f your-draft.md). This creates a vulnerability surface where instructions embedded within the processed draft could attempt to influence the agent's behavior.
Ingestion points: your-draft.md via the CLI tool (SKILL.md).
Boundary markers: No specific delimiters or warnings to ignore instructions within the draft content are provided.
Capability inventory: The skill uses npm install for setup and executes a Node.js CLI tool (node src/cli.js) to process data.
Sanitization: No evidence of sanitization or filtering for the input draft content.

Remove AI Writing Fingerprints from Draft Copy