MCP skill registry for writing styles — unlimited skills, live from GitHub, no re-installs.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"writing-skills-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# writing-skills-mcp **The living registry for [writing-skills](https://github.com/surendranb/writing-skills): unlimited writing styles, served live from GitHub, installed on demand.** The [writing-skills](https://github.com/surendranb/writing-skills) repo ships 14 pristine SKILL.md files with zero telemetry. This server is the optional, separate layer on top: it discovers styles from the full catalog at call time, so a new skill is one commit away from every user — no package update, ever. ## Install ```json { "mcpServers": { "writing-skills": { "command": "uvx", "args": ["writing-skills-mcp"] } } } ``` ## Tools | Tool | Does | | :--- | :--- | | `search_styles` | Ranked match over the full style catalog | | `list_styles` | All available style slugs | | `get_skill` | Full SKILL.md for any style, fetched live | | `install_skill` | Writes a style into your harness skills directory | ## Remote-first roadmap The catalog and skill content are already remote (fetched from GitHub `main` at call time; the bundled catalog is the offline fallback). Next: a hosted streamable-HTTP endpoint so hosted harnesses (claude.ai, Cowork) can connect without any local install — and so every user always runs today's server. ## Telemetry This server (unlike the skills repo, which has none) reports anonymous usage — boots, tool calls, errors — to find where it falls short. No content, no arguments, no PII. Opt out absolutely: `WRITING_SKILLS_TELEMETRY=false`, `DO_NOT_TRACK`, or `NO_TELEMETRY`. ## License MIT — [Surendran B](https://github.com/surendranb)
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.