Skip to content

v1.0.4 — Honesty Audit

Choose a tag to compare

View all tags
@devag7 devag7 released this 13 Jun 07:01
· 44 commits to main since this release
v1.0.4
0fbdb5a

Documentation Integrity Release

Live testing confirmed all data tools fail against LinkedIn due to Cloudflare bot-management (302 redirect loops that plain HTTP cannot resolve — verified with both Node fetch and curl).

This release corrects every false/misleading claim:

What Changed

  • README: Removed "most reliable" tagline. Added ⚠️ Cloudflare limitation warning. Added "Known Limitations" section. Rewrote comparison table truthfully.
  • AGENTS.md: Added limitation warning. Fixed OAuth claim (0 call sites, not "experimental").
  • mcp-manifest.json: Corrected both model and human descriptions.
  • package.json: Honest description with Cloudflare note.
  • whoami tool: Capabilities now include note about Cloudflare blocking.
  • health_check tool: Description clarifies it checks linkedin.com reachability, NOT Voyager API access.

What Still Works

  • MCP protocol (36 tools register, schemas valid) ✅
  • whoami, health_check, CLI ✅
  • Cookie jar + redirect handling (v1.0.3) ✅
  • 53 tests, all passing ✅

Install

npx -y linkedin-mcp-tools@1.0.4

Full Changelog: v1.0.3...v1.0.4

Assets 2
Loading