deploy-doctor
Skill0 starsChanged 38 days agoMITView on GitHub
Diagnoses and fixes a failed or broken Vercel deployment in plain English. Use when a build or deploy fails, the live site errors or is down, or someone asks 'why won't it deploy'.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Yahiasherif002/claude-skillstheir own version, first seen Jun 26, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- Yahiasherif002/claude-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- access-checkSkill
- data-checkSkill
- domain-helperSkill
- first-tenSkill
- idea-to-mvpSkill
- keys-checklistSkill
- make-it-yoursSkill
- payment-testSkill
- project-starterSkill
- review-loopSkill
- safe-changeSkill
- sketch-itSkill
- stack-checkSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Read its build logs (and runtime logs if the build passed but the site errors).
- Distinctive: line 13
Check whether the live site is still on the last working version.
- Distinctive: line 20
No stack traces unless they ask.
- Distinctive: line 31
Only merge to main (which deploys live) when they say so.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Deploy doctor
You're helping someone who doesn't code fix a deployment. Diagnose first, explain plainly, then fix safely.
1. Find the problem
- Using the Vercel connector, find the project and its most recent deployments. Identify the failed one (or the latest one if the site is live but broken).
- Read its build logs (and runtime logs if the build passed but the site errors). Find the first real error, not the noise after it.
- Check whether the live site is still on the last working version. (Vercel keeps the last good deployment live when a build fails, so visitors usually aren't affected. Confirm this and reassure them if so.)
2. Explain it
In one short paragraph, plain English:
- What broke: e.g. "The build stopped because a page imports a file that doesn't exist."
- Why: e.g. "It was renamed in the last change, but one page still uses the old name."
- Is the live site affected? yes or no. No stack traces unless they ask.
3. Common causes to check
- A missing environment variable (a key set on the computer but not in Vercel). Say which name is missing and where to add it in Vercel's settings. Never ask for the key's value, and never put it in code. They add it themselves.
- A typo, a renamed or missing file, or a missing package.
- A type or lint error that only fails on the build server.
- A database or API the app can't reach from Vercel.
4. Fix it safely
- Make the fix on a branch, push it, and use the preview deployment to confirm the build passes.
- Tell them what you changed in plain English and share the preview link.
- Only merge to
main(which deploys live) when they say so.
5. If the live site is broken right now
Offer the fast, safe option first: roll back to the previous working deployment in Vercel (one or two clicks, or via the connector if they say yes), then fix calmly. Note: after a rollback, new pushes don't go live automatically until the rollback is undone. Remind them to undo it once the fix is ready.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.