tool-connector
Skill472 starsChanged 8 months agoMITView on GitHub
Connect any tool you use at work to your agent — including internal company tools, custom-built systems, deployment portals, incident trackers, internal knowledge bases, HR systems, and commercial tools like Slack, Confluence, Jira, Linear, GitHub, Outlook, Datadog, and PagerDuty. Use when the user wants to set up a tool connection, connect an internal or custom-built tool, or add a new tool integration from scratch.
In this repository11 other files it gives its agents.
- assistant-orchestratorSkill
- colleague-distillationSkill
- create-workflowSkill
- discover-ui-surfaceSkill
- linkedin-engagementSkill
- assistant-orchestratorSkill
- colleague-distillationSkill
- create-workflowSkill
- discover-ui-surfaceSkill
- enterprise-searchSkill
- linkedin-engagementSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
If a tool has an API or a web UI, it can be connected.
- Very specific: line 36
Never edit tool_connections/ directly — copy to personal/{tool}/ and patch there.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Tool Connector — Connect Any Work Tool to Your Agent
Requires the repo cloned locally. Check if it's already present (
ls setup.md), then readsetup.md— that is the full entry point. The sections below summarize the approach for reference.
Full methodology and pre-built recipes: https://github.com/zhixiangluo/10xProductivity
Internal tools are first-class citizens
Internal and custom-built tools are often the most valuable to connect — deployment portals, incident trackers, internal knowledge bases, custom dashboards, HR systems, anything your company built or runs. They follow the exact same setup path as commercial tools and stay private in personal/ (gitignored, never committed).
If a tool has an API or a web UI, it can be connected.
Core principles
- Ask for a URL first, not credentials — a URL the user has open reveals the base URL, workspace, and regional variant without requiring them to know anything about auth
- Run every command yourself — never paste a command and ask the user to run it
- Validate with real output before marking anything done — no illustrative output, no copy-paste from docs
- Zero friction — never require the user to create an OAuth app, register redirect URIs, or configure anything outside of this flow
Routing
| Situation | Action |
Tool already in verified_connections.md | Reverify — run its verify snippet; if it passes, done |
Tool has a recipe in personal/{tool}/ | Load it and try; patch in personal/ if it fails |
Tool has a recipe in tool_connections/{tool}/ | Read that tool's setup.md and follow it |
| Tool not found anywhere | Follow add-new-tool.md to build a recipe from scratch |
Never edit tool_connections/ directly — copy to personal/{tool}/ and patch there.
Auth priority order
| Priority | Method | User friction |
| 1 | API token | ~30s — generate in tool settings |
| 2 | Browser session (SSO, cached) | Run once — session cached for days/weeks |
| 3 | Browser session (per operation) | Playwright on every call — only if no API exists |
| 4 | Username + password | Legacy tools only |
| ✗ | OAuth requiring user to create their own app | Never — too much friction |
Pre-built recipes available
Slack, Confluence, Jira, Linear, GitHub, Microsoft Teams, Outlook, Google Drive, Datadog, Grafana, PagerDuty, Jenkins, Backstage, Bitbucket Server, Artifactory.
Internal and custom tools follow the same path — they stay private in personal/ (gitignored).
Getting started
First, check if you already have the repo:
ls setup.md 2>/dev/null && echo "repo present" || echo "need to clone"Repo present → if python3 (or Windows py -3) is missing or .venv is not set up for Playwright, read setup-python.md first; then read setup.md — it is the full entry point for tool connections.
Need to clone →
git clone https://github.com/zhixiangluo/10xProductivity.git
cd 10xProductivityThen read setup.md. It routes to pre-built recipes, your own recipes, and add-new-tool.md for anything new.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

