Humanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 24 pattern detectors, 500+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content,
Security Analysis
high confidenceThe skill's code, runtime instructions, and resource requirements are consistent with its stated purpose (detecting and humanizing AI‑style writing); it asks for no credentials and contains no obvious external endpoints or install steps.
Name/description (humanize/de-AI text) align with the included code (pattern detectors, vocabulary lists, stats engine). The declared requirements are minimal (no env vars, no binaries) which matches a pure Node.js analysis tool. Nothing requested appears unrelated to the stated purpose.
SKILL.md instructs the agent to scan and rewrite user-provided text for 24 defined patterns and to use statistical signals — it does not instruct reading unrelated files, trampling system config, or exfiltrating data. Examples and CLI usage are local/offline oriented.
There is no install spec (instruction-only at registry level) while full source code is bundled. The codebase is pure Node.js with no build or remote-download install steps. No URLs that download arbitrary/executable archives are used in the skill metadata.
The skill declares no required environment variables, credentials, or config paths. The code and scripts shown do not reference external secrets or services; all pattern detection, vocabulary, and stats run locally.
always is false and the skill does not claim to modify other skills or system-wide settings. It does not request elevated or persistent privileges.
Guidance
This skill is internally coherent: the code implements the 24 pattern detectors and statistical checks described in SKILL.md, it runs locally and requests no credentials, and there are no obvious network or install steps. If you plan to use it on sensitive text, note that it processes content locally (no APIs required) according to the README — verify the runtime environment (Node >=18) before running the bundled scripts. If provenance/trust matters, consider reviewing the full source files (src/*.js and references/*.md) or using the package from the upstream GitHub repo listed in package.json to confirm the author and recent commits.
Latest Release
v2.1.0
ESLint v9 flat config, all lint errors fixed, Prettier formatted, shared config framework
More by @brandonwise
Published by @brandonwise on ClawHub
