install
Skill83 starsChanged 3 days agoApache-2.0View on GitHub
Install, update, clean up, uninstall, or check the status of openevidence-mcp on this machine. Use when the user wants to set up openevidence-mcp, update to the latest release, register/unregister the MCP server with Claude/Codex, reap stray relay daemons, or verify the install. Trigger phrases include "install openevidence-mcp", "update openevidence-mcp", "uninstall openevidence-mcp", "openevidence-mcp 更新 / 安裝 / 移除 / 清理", "reap relay daemon", "is openevidence-mcp registered".
23 other projects use a version of this.
Adopted by
Most recently taken up by
- RLinf/RLinftheir own version, first seen Nov 15, 2026
- Dariusplayer09/adaptive-skillstheir own version, first seen Nov 8, 2026
- kkrafft1999/snotratheir own version, first seen Nov 6, 2026
- tykimos/agent-native-agenttheir own version, first seen Nov 6, 2026
- crocotaste/crocotaste-plugintheir own version, first seen Nov 1, 2026
- BrainWeb/payload-mcp-oauththeir own version, first seen Nov 1, 2026
Used with
Projects with it also use update and release.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
AnthropicSkillupdate
Sync tasks and refresh memory from your current activity. Use when pulling new assignments from your project tracker into TASKS.md, triaging stale or overdue tasks, filling memory gaps for unknown people or projects, or running a comprehensive scan to catch todos buried in chat and email.
Also in 3 projects that use this
MicrosoftSkillrelease
Use ONLY when the user explicitly asks to cut/publish a Huabu desktop release or tag a version. Performs an outward action — bumps the version in apps/desktop/package.json, commits + annotated-tags vX.Y.Z, and pushes the branch + tag. That tag push auto-builds the macOS/Windows installers and publishes a GitHub Release. Provides `status`, `bump`, `tag-push`, and a one-command `run` (bump → tag-push).
Also in 3 projects that use this
Other versions23 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 23
- kubeshark/kubeshark★ 12ksame name
- RLinf/RLinf★ 5.7ksame name
- PrimeIntellect-ai/prime-rl★ 2.3ksame name
- tykimos/agent-native-agent★ 147same name
- jeremylongshore/claude-code-slack-channel★ 43same name
- BrainWeb/payload-mcp-oauth★ 9same name
- kkrafft1999/snotra★ 6same name
- WeDataLab-AI-Lab/cctrace★ 5same name
- gpambrozio/Gallager★ 4same name
- Pierry/harness-kit★ 4same name
13 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository4 other files it gives its agents.
- docs/llms-full.txtllms.txt
- docs/llms.txtllms.txt
- openevidence-rate-limitsSkill
- openevidenceSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 0 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 19 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 16
Run the subcommand that matches the user's intent, then read the printed guidance back to them — especially the reload the browser …
- Very specific: line 37
If $CLAUDE_SKILL_DIR is not set, use the repo path directly: .claude/skills/install/scripts/oe-mcp.sh <command>.
- Very specific: line 80
Only the full-wipe commands printed by uninstall are destructive — get explicit confirmation before running those.
“Common” is measured: kept by at least 6 in 10 of the 19 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
openevidence-mcp — install / update / cleanup / uninstall
One skill to set up, update, and tear down openevidence-mcp
on this machine. Everything here wraps the repo Makefile, so it stays in sync with
the canonical install flow — the skill never reimplements it.
The single entry point is scripts/oe-mcp.sh (a thin dispatcher over make).
Run the subcommand that matches the user's intent, then read the printed guidance
back to them — especially the reload the browser extension step, which is a
chrome:// GUI action that cannot be scripted.
Pick the action
| User wants… | Run |
| First-time setup | scripts/oe-mcp.sh install |
| Update to the latest release | scripts/oe-mcp.sh update |
| Check it's working | scripts/oe-mcp.sh status |
| Relay should survive reboots / "start at boot" (macOS) | scripts/oe-mcp.sh autostart |
| Reap stray daemons / free stale state | scripts/oe-mcp.sh cleanup |
| Remove it | scripts/oe-mcp.sh uninstall |
Invoke the script by its path from the skill directory, e.g.:
bash "$CLAUDE_SKILL_DIR/scripts/oe-mcp.sh" updateIf $CLAUDE_SKILL_DIR is not set, use the repo path directly:
.claude/skills/install/scripts/oe-mcp.sh <command>.
What each action does (all via make)
- install —
make all: installs deps, builds the MCP server (dist/server.js) and the relay extension (extension/dist), and registers the server into Claude and Codex (whichever CLI is present). If run outside a checkout, it clones the repo to$HOME/openevidence-mcpfirst (override withOE_MCP_DIR). - update —
make update:git pullthe latest release, rebuild server + extension, re-register. Prints the extension-reload reminder. - status —
make status: server + extension versions, live relay/health, and whether the CLIs have it registered. - autostart —
make autostart: installs a launchd LaunchAgent so the relay daemon starts at login, restarts on crash and never idle-reaps. Undo withmake autostart-remove. The browser (extension + logged-in tab) must still be running — suggest adding it to macOS Login Items. - cleanup —
make cleanup: reaps orphan relay daemons (keeps the live one), prunes oversized relay logs and temp answer artifacts. Non-destructive — your install, registration, and~/.openevidence-mcpdata are untouched. The relay auto-respawns on the next MCP call. - uninstall —
make uninstall: unregisters from the AI CLIs, stops daemons, removesdist/. Deliberately keeps~/.openevidence-mcp(your SQLite db + config) and the loaded browser extension — the skill should tell the user the two manual commands to fully wipe those, and only run them if explicitly asked.
The one manual step (install & update)
Loading/reloading the browser extension is a chrome://extensions GUI action —
you cannot do it for the user. After install/update, tell them exactly:
- Open
chrome://extensions(Chrome / Edge / Brave / Arc / Vivaldi / Opera). - Turn on Developer mode. First time: Load unpacked → select
<repo>/extension/dist. After an update: click Reload on "OpenEvidence MCP Relay". - Stay logged in to openevidence.com in that browser.
- Reconnect
/mcpin the AI client.
Then verify: curl -s http://127.0.0.1:8787/health → expect "connected":true.
Notes
- These are host-level, reversible operations; run the matching command directly.
Only the full-wipe commands printed by
uninstallare destructive — get explicit confirmation before running those. - The skill is host-specific: it acts on the machine it runs on. Nothing here touches the user's OpenEvidence account or data on the server side.
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.

