Writing Style Checker

July 14, 2026

Version updated for https://github.com/theserverlessdev/wsc to version v1.1.0.

This action is used across all versions by 0 repositories.

Action Type

This is a Composite action.

Go to the GitHub Marketplace to find the latest changes.

Action Summary

WSC is a GitHub Action designed to automate the detection of AI-generated text patterns, including problematic writing styles such as weasel words, passive voice, duplicate words, long sentences, nominalizations, hedging, filler adverbs, and AI-specific constructs. It offers real-time feedback in a web editor, can be accessed via an API, integrated into MCP servers, run as a CLI, and is available as a GitHub Action. WSC helps improve writing quality by identifying and correcting common issues without relying on authorship proof.

What’s Changed

fix(cli,tests): force color via FORCE_COLOR; ANSI test no longer depends on module identity (3240491)
feat(action): move composite action to repo root for Marketplace listing (558cef7)
fix(core,web,mcp,cli): strip mask sentinels from aiTells text, correct stale counts, deterministic ANSI test (2d54f33)
chore(release): wsc-lint 1.3.0, wsc-mcp 2.2.0 (b559496)
feat(site,docs): AI-tells-first positioning, honest-detection statement, scoped privacy claim, freelance credit line (554b19b)
fix(docs,meta): correct Action uses: path, drop phantom MCP tool and dead ai-plugin.json, complete npm metadata (573aae9)
feat(cli): colored output via picocolors, working –no-color, version from package.json, reject unknown formats (0274017)
feat(web): persisted Markdown-mode toggle in the editor (b597be5)
feat(api,mcp): markdown-aware analysis on every surface; fix MCP version drift (7e708f5)
feat(core): Markdown masking moves into analyzeText with sentinel-based inline masking (d1be05c)