LiGo RunnerSAFE
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
The first GPT-powered creative co-pilot trained on your actual LinkedIn content.
Create posts. Analyze what’s working. Rewrite in your voice. All from Claude or ChatGPT, powered by your own past posts.
What is this?
This is the official runner repo for LiGo’s Model Context Protocol (MCP) - a protocol that lets GPT-based assistants pull your LinkedIn context and respond like a strategist. It works with both Claude and ChatGPT.
With MCP, your assistant can answer things like:
- Which of your last posts got the most traction
- What your writing tone actually sounds like
- How to help you post, rewrite, or brainstorm like a founder
Just talk to it normally (if you know how to do that - no judging if you don't, btw). It has access to all your PUBLIC Linkedin data (with your consent, of course).
How to Get Started
Claude Setup
- Download the Claude desktop app.
- Visit ligo.ertiqah.com/integrations/claude
- Click "Generate Installation Command"
- If not logged in, you’ll be routed to authenticate with LiGo
- Copy the command and run it in your terminal
- Open Claude and start chatting
Example prompt:
Analyze my last 5 posts. What’s working? Give me ideas on what I should write next.
ChatGPT (CustomGPT)
No installation needed.
- Go to ligo.ertiqah.com/integrations/chatgpt
- Authenticate with LiGo when prompted
- Start using the CustomGPT
Example prompt:
Rewrite this to sound more like my recent posts and make the hook spicier.
##
929bae80a21aOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add linkedin-mcp-runner --env LINKEDIN_MCP_API_KEY=${LINKEDIN_MCP_API_KEY} -- npx -y [email protected]{
"mcpServers": {
"linkedin-mcp-runner": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"LINKEDIN_MCP_API_KEY": "${LINKEDIN_MCP_API_KEY}"
}
}
}
}Exposed tools (10)
3 read · 5 write · 2 destructive. Blast radius: 2 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
analyze_linkedin_chat | read | Ask questions about the user |
generate_linkedin_post | write | Generate three LinkedIn post variants from any content (article, newsletter, notes, etc.) to optimize engagement. |
get_linkedin_posts | read | Retrieve the user |
get_linkedin_profile | read | Retrieve the user |
publish_linkedin_post | write | Publish a text post to LinkedIn, optionally including media (images/videos) specified by URL. |
publish_twitter_post | write | Publish a text post (tweet) to Twitter. |
refresh_linkedin_posts | destructive | Force a refresh of LinkedIn posts data to capture recently published content. |
refresh_linkedin_profile | destructive | Force a refresh of the LinkedIn profile data to update any recent changes. |
schedule_linkedin_post | write | Schedule a text post for LinkedIn at a specific future date and time, optionally including media (images/videos) specified by URL. |
set_linkedin_url | write | Set or update the LinkedIn profile URL to analyze. Required before using profile/posts retrieval tools if not set previously. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- UNDECLARED (1 observation(s))
- Network
- declared (5 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (2)
refresh_linkedin_posts, refresh_linkedin_profile
axios, fs-extra
Gates applied: no_behavioural_pass.
929bae80a21afull audit observations/trust-audit/mcp-server/ertiqah__ligo-runner.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 929bae80a21a | SAFE | B | 89 | first audit |
Questions
What tools does LiGo Runner expose?
10 in total: 3 read-only, 5 that write, and 2 that can delete or overwrite (refresh_linkedin_posts, refresh_linkedin_profile). Every one is listed on this page with its risk.
Is LiGo Runner safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 2 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does LiGo Runner need?
It reads LINKEDIN_MCP_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How current is this page?
The grade is for one exact copy of the source (929bae80a21a), read on 2026-10-08. The repository is watched and re-audited when it changes.