Borrowing it
Nothing to install: this file belongs to larsbaunwall/Unlinked. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/larsbaunwall/Unlinked/main/AGENTS.mdgit clone --depth 1 https://github.com/larsbaunwall/UnlinkedWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/larsbaunwall/unlinked/agents-md)<a href="https://agentmods.dev/instructions/larsbaunwall/unlinked/agents-md"><img src="https://agentmods.dev/badge/instructions/larsbaunwall/unlinked/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/larsbaunwall/unlinked/agents-md"><img src="https://agentmods.dev/badge/instructions/larsbaunwall/unlinked/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02615 | $0.02615 |
| Opus 5.5 | $0.01046 | $0.01046 |
| Sonnet 5.5 | $0.00523 | $0.00523 |
| Haiku 4.5 | $0.00262 | $0.00262 |
Grade A, and why
Unlinked AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project Purpose
This repository is Unlinked, a TypeScript MCP server that connects a user's LinkedIn professional profile and experience data to AI assistants such as Claude Desktop, GitHub Copilot, and other Model Context Protocol clients.
Unlinked exists to let a member bring their own professional context into an assistant without making the assistant scrape LinkedIn, infer profile details, or rely on stale manually copied text. The LinkedIn API surface used here is read-only.
Unlinked runs in two modes from one binary: a JSON CLI (the default: unlinked profile, section, activity, status, cache clear) and an MCP server (unlinked --mcp). Both are thin front ends over src/service.ts; keep behaviour in the service, not in the adapters (src/cli.ts, src/tools/linkedin.ts).
Core Requirements
- Build the server with the official TypeScript MCP SDK,
@modelcontextprotocol/sdk. - Run the MCP server over stdio with
StdioServerTransportwhen started with--mcp. Any other invocation is the CLI. - Keep the project usable by desktop MCP clients, especially Claude Desktop and GitHub Copilot.
- Connect to LinkedIn's Member Data Portability (Member) API product.
- Fetch LinkedIn member profile and professional-history data for the authenticated member.
- Treat the LinkedIn access token as a startup secret. Read it from the
LINKEDIN_TOKENenvironment variable. Use it only to sendAuthorization: Bearer <access_token>to LinkedIn. Never persist it, log it, echo it into MCP responses or CLI output, or put it in package scripts. The cache directory name uses only a truncated sha256 of the token. - Bounded cache exception to "do not persist". LinkedIn data is cached to protect the API: in memory for MCP, in memory plus disk for the CLI (
<OS cache dir>/unlinked/<sha256(token)[0:16]>/, directory 0700, files 0600, atomic writes), 6 h TTL by default (UNLINKED_CACHE_TTL,0= off), empty results at most 10 minutes. Only sanitized data is ever cached: snapshot rows afternormalizeRowredaction, and changelog events after droppingmessages/invitationsand keeping only the fields the merge and activity feed use. Blocked domains are never fetched. Errors are never cached. Tests assert that no cache file contains a redacted key, the token, or a LinkedIn domain name. - Names match the LinkedIn UI. LinkedIn's internal domain names (
POSITIONS,ALL_LIKES) live only insrc/linkedin/sections.tsand never appear in tool descriptions or output. - Sensitive domains (INBOX, LOGIN, PHONE_NUMBERS, EMAIL_ADDRESSES, CONTACTS, ads, searches, receipts, ...) are blocked in
sections.tsand must never be fetched or exposed. Private fields inside exposed sections (birth date, address, contact emails/phones, ...) are redacted. - Make clear in user-facing docs and errors that the Member Data Portability product is currently available only to LinkedIn members in the European Economic Area and Switzerland.
- Preserve the API's read-only nature. Do not add tools that mutate LinkedIn data.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · +28 lines · +1,460 tokens per session 825a5909fbfa
- 2mo ago First seen · 63 lines · 1,155 tokens per session scan A 1dcebbe89e92
Unlinked AGENTS.md is an instructions file published in the GitHub repository larsbaunwall/Unlinked (4 stars, last pushed 1mo ago), licensed MIT. It adds 2,615 tokens to every session, about $0.0105 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Unlinked AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| Unlinked AGENTS.mdlarsbaunwall/Unlinked | Instructions file | 4 | 1mo ago | MIT | A | 2,615 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |