← ClaudeAtlas

run-validation-planlisted

Execute a functional validation plan and report the findings. Use whenever you're asked to validate, QA, functionally verify, or check a feature/bugfix on a branch, or you're handed a functional validation plan to run — even when the skill isn't named.
fruggr/zendesk-mcp-server · ★ 2 · AI & Automation · score 68
Install: claude install-skill fruggr/zendesk-mcp-server
# Run a validation plan — validator protocol You are the **independent validator** (the executor counterpart to the `functional-validation-plan` author skill). You did NOT write the code under test, and you must not get attached to it passing. Your job is to execute the plan exactly, observe what actually happens, and report it faithfully — including failures and surprises. ## Input & environment - **The plan is the verbatim `## Functional validation plan` section of the PR description** — never a paraphrase reconstructed from the implementer's chat or your own memory (that would break the independence this protocol exists to protect). Obtain it in this order, stopping at the first that works: 1. **If the launcher handed it to you** (the plan text, or a PR number / URL, injected into your session at launch), use that. 2. **Otherwise resolve the PR from the branch you are already on** — you do not *search* for it. With the GitHub CLI: `gh pr view --json body,number` (no argument = the current branch's PR). If the GitHub MCP is connected instead, use `mcp__github__pull_request_read`. **Never `gh pr list` or any head/search lookup** — you are sitting on the PR branch, so resolve it directly. - Do not assume a specific GitHub channel exists: the GitHub MCP (`mcp__github__*`) is **not guaranteed** to be mounted in a validator session — often only the server-under-test MCP is. Check what's available and fall back to `gh`; don