3-verify-pr-fix

Verifies whether a candidate PR fixes a reproduced LinkedIn MCP server issue by replaying the failing tool call.

Updated Aug 20, 2026
One-click install
npx skills add https://github.com/abetoluwani/linkedin-mcp-server --skill 3-verify-pr-fix-abetoluwani
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: 3-verify-pr-fix
Source: https://github.com/abetoluwani/linkedin-mcp-server/tree/main/.agents/skills/3-verify-pr-fix
Command: npx skills add https://github.com/abetoluwani/linkedin-mcp-server --skill 3-verify-pr-fix-abetoluwani

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When reviewing pull requests for the LinkedIn MCP server, maintainers need objective proof that a PR actually fixes a reported issue rather than just plausible-looking code. This Skill checks out the PR locally, restarts the MCP server on the PR branch, replays the exact MCP tool call that failed on main, diffs the outputs, and audits the diff against scraping rules to produce a clear verdict. ## Core Features & Use Cases - Baseline Replay: Reuses the on-main baseline and tool-call metadata captured by /2-repro-issue to re-run the identical MCP call against the PR branch over streamable HTTP. - Verdict Classification: Classifies the PR as fixes cleanly, fixes with concerns, does not fix, cannot run, or cannot verify due to skill limitations, with evidence from the output diff. - Scraping Rule Audit: Flags locale-dependent detection, LinkedIn class-name selectors, multi-navigation section mappings, and missing tests per CLAUDE.md rules. - Use Case: A maintainer asks "does PR #498 fix issue #470?" — the Skill checks out the PR, replays the failing get_person_profile call, diffs the JSON responses, and reports a verdict with merge recommendation. ## Quick Start Ask the assistant to verify PR #498 against the issue it fixes, after running /2-repro-issue to capture the on-main baseline.

Frequently Asked Questions about 3-verify-pr-fix

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify a PR fixes a GitHub issue locally?▼

Check out the PR head as a detached HEAD, restart the MCP server on a free port, replay the exact tool call captured in the baseline meta file, and diff the JSON output against the on-main baseline. This Skill automates that flow and reports a verdict.

How to test an MCP server tool call over HTTP?▼

Start the server with streamable-http transport, POST an initialize request to /mcp to obtain an Mcp-Session-Id header, send notifications/initialized, then POST a tools/call JSON-RPC request with the tool name and arguments.

What happens if there is no reproduction baseline for the issue?▼

The Skill stops and instructs you to run /2-repro-issue first, since that skill is the canonical source of the failing tool call and on-main response. It never silently re-runs the reproduction itself.

Can this verify crash-only or stdio-only bugs?▼

No. If the baseline never captured a failure, the verdict is reported as cannot verify fix due to skill limitation, and the Skill falls back to a regression check plus a code audit rather than criticizing the PR.

Why does the audit flag locale-dependent string checks in scraping code?▼

LinkedIn UI text like Connect or Pending changes with user locale, so equality checks on translated verbs break detection for non-English profiles. Attribute presence checks are locale-independent and preferred per CLAUDE.md rules.