Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/devag7/linkedin-mcp/agents-mdgit clone --depth 1 https://github.com/devag7/linkedin-mcpWhat 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 | $0.00582 | $0.00582 |
| Opus 5 | $0.00291 | $0.00291 |
| Sonnet 5 | $0.00116 | $0.00116 |
| Haiku 4.5 | $0.00058 | $0.00058 |
Grade A, and why
linkedin-mcp 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 2d 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI agents using or contributing to LinkedIn MCP
(linkedin-mcp-tools).
What this is
An MCP server exposing LinkedIn to any MCP client (Claude, Cursor, …) as structured JSON. It drives a real stealth Chrome to pass Cloudflare, then calls LinkedIn's Voyager API from inside the authenticated page. 22 tools.
Using it (as an agent / MCP client)
- The human runs the one-time login:
npx -y linkedin-mcp-tools@latest --login. - Configure the MCP server: command
npx, args["-y","linkedin-mcp-tools@latest"]. - Call tools. Reads return structured JSON. Writes require
confirm: trueand count against daily safety caps — only call them when the user explicitly asks, and surface the returnedstatus(ok/duplicate/already_connected/restricted/quota_exhausted/not_allowed/failed) back to the user. - Check
health_checkif a call fails — it reports login state, a live API probe, and remaining daily budget.
Tool map
- Profiles:
get_my_profile,get_profile - Search:
search_people,search_jobs,get_job_details,search_companies,get_company,get_company_posts,get_company_employees - Feed/messaging:
get_feed,get_notifications,get_inbox,get_conversation,get_pending_invitations - Writes (gated):
connect_with_person,send_message,create_post,react_to_post,comment_on_post - Session:
whoami,health_check,close_session
Safety expectations
Automating LinkedIn can get an account restricted — there is no ban-proof tool.
Default to a secondary account, respect the built-in daily caps, never disable
pacing for real use, and stop immediately if health_check reports a checkpoint.
Contributing (code agents)
- TypeScript, strict.
npm run build(tsup),npm run typecheck,npm test(vitest),npm run lint(eslint) — all must pass; CI enforces them. - Voyager endpoints rotate. Write endpoint shapes are captured live via
--writecapture(intercept the SPA's POST, abort it — zero side effects) and verified via--writeprobe; do not guess payloads. - Version is single-sourced from
package.json(src/version.ts). Releases are automated on push tomainwhen the version bumps (see.github/workflows).
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.
- 2d ago First seen · 49 lines · 582 tokens per session scan A 07abe4be1a9f
linkedin-mcp AGENTS.md is an instructions file published in the GitHub repository devag7/linkedin-mcp (9 stars, last pushed 5d ago), licensed MIT. It adds 582 tokens to every session, about $0.0029 per session on Opus 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.
Other instructions, from other repositories
qaskills AGENTS.md
Instructions for PramodDutta/qaskills, covering agents.md, project overview, common commands, run a single package and run tests in watch mode.
agentic-playwright ai-native-workflow.instructions.md
Instructions for idavidov13/agentic-playwright, covering ai-native workflow, critical, main workflow (8 phases), phase 4 — confidence-gate format (mandatory) and proposal.
agentic-playwright data-strategy.instructions.md
Instructions for idavidov13/agentic-playwright, covering data strategy, critical, file locations, instructions and phase 1: classify the value you need.
agentic-playwright config.instructions.md
Instructions for idavidov13/agentic-playwright, covering configuration, critical, file locations, instructions and phase 1: understand how env files are loaded.
agentic-playwright refactor-values.instructions.md
Instructions for idavidov13/agentic-playwright, covering refactoring enum values and static test data, critical, instructions, phase 1: find all consumers before touching anything and find all usages of the enum key.
inspecto CLAUDE.md
Instructions for inspecto-dev/inspecto, covering inspecto — claude code development guide, project overview, monorepo structure, development phases (load each file as needed) and key architectural decisions.