CtrlK
BlogDocsLog inGet started
Tessl Logo

connect-openwork-mcp

Connect the OpenWork MCP Gateway (https://api.openworklabs.com/mcp/agent) to Claude Code, Codex, Gemini CLI, Cursor, VS Code, Claude Desktop, or ChatGPT so the agent can use the user's OpenWork organization skills, plugins, and connections.

68

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exemplary lean operational guide: exact commands for the main clients, an explicit sign-in step, a verification gate, and targeted troubleshooting with recovery loops. The only notable gap is that six of the ten supported clients are handled by a generic instruction plus an external link rather than inline steps.

DimensionReasoningScore

Conciseness

"Server URL: ... Transport: Streamable HTTP. Auth: OAuth (PKCE, dynamic client registration)" and the terse per-client command blocks contain no concept explanations Claude already knows and no padding — every line is operational. Matches anchor 5: lean, assumes competence, every token earns its place.

5 / 5

Actionability

Copy-paste-ready commands are given for Claude Code, Codex, Gemini CLI, and OpenCode (e.g. `claude mcp add --transport http openwork https://...`), with concrete sign-in and verification steps. The six remaining clients ("Cursor, VS Code, Claude Desktop, ChatGPT, Windsurf, Zed: add ... as a remote MCP server") get only a high-level instruction plus a docs URL rather than exact steps — a minor gap that fits anchor 4 rather than 5.

4 / 5

Workflow Clarity

A clear three-stage sequence (Add → Sign in → Verify) with explicit validation checkpoints ("The agent has the search_capabilities and execute_capability tools") and a hard gate: "Don't claim the connection works until step 3 passes." The failure section adds a feedback loop (401/invalid_grant → log out and sign in again), matching the top anchor.

5 / 5

Progressive Disclosure

Sections are well organized and content is appropriately placed: common clients inline with exact commands, niche clients deferred via a clearly signaled one-level-deep link ("Per-client steps: https://openworklabs.com/docs/start-here/connect-openwork-mcp"), plus a clearly labeled reference link. There is no bundled-file sprawl and no nested reference chains.

5 / 5

Total

19

/

20

Passed

Description

70%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concrete, specific, and highly distinct, with excellent natural trigger terms in the client names. Its main weakness is the missing 'when should this be used' trigger clause, which caps completeness and leaves some natural phrasings uncovered.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to connect OpenWork, add the OpenWork MCP server/gateway, or wants their OpenWork organization skills available in their agent client."

Include common synonyms such as "MCP server", "set up OpenWork", or "install OpenWork" alongside "MCP Gateway" to broaden natural-language matching.

Mention what the user gains functionally after connecting (e.g. searching and executing organization capabilities) to sharpen the 'what' further.

DimensionReasoningScore

Specificity

"Connect the OpenWork MCP Gateway (https://api.openworklabs.com/mcp/agent) to Claude Code, Codex, Gemini CLI, Cursor, VS Code, Claude Desktop, or ChatGPT" names one concrete action across an enumerated list of specific targets, with an explicit outcome ("use the user's OpenWork organization skills, plugins, and connections"). Matches anchor 4 (specific actions, minor coverage gaps) rather than 5 (only one action verb) and well above anchor 3's '1-2 concrete actions'.

4 / 5

Completeness

The 'what' is clear (connect the gateway to a specific set of clients so the agent can use organization skills), but the 'when' is entirely absent — there is no 'Use when...' clause or equivalent explicit trigger guidance. Per the judging guidelines, a missing 'Use when...' clause caps completeness at 3.

3 / 5

Trigger Term Quality

The client names ("Claude Code, Codex, Gemini CLI, Cursor, VS Code, Claude Desktop, ChatGPT") plus "OpenWork" and "MCP Gateway" are exactly the natural phrases a user would say when asking to connect OpenWork. A few common variations are missing (e.g. "MCP server", "set up", "install"), so it does not reach anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

"OpenWork MCP Gateway" is a clear, branded niche with distinct triggers; it is unlikely to fire for the wrong skill since both the product name and the enumerated client list are unambiguous.

5 / 5

Total

16

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
different-ai/openwork
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.