convex-launch-readiness Skill for Claude Code: What It Does and How to Install It

Makes Claude run Convex's audits together and merge them into one scored report with an ordered fix plan.

Installs: 38.3K; on skills.sh
Stars: 68; on GitHub
Updated: Oct 2026; last repo commit

This skill makes Claude run the Convex audit skills together and turn their results into one report. It runs the authorization scan, the code review, the performance check and the logs check, removes duplicates and gives a score from 0 to 100. The report lists findings as blockers, should-fix and nice-to-have, then an ordered fix plan. SKILL.md describes it as "Lighthouse for your backend".

The score starts at 100 and loses 15 per confirmed high finding, 5 per medium and 1 per low. It prints the formula so the number can be reproduced. Passes that cannot run, such as the performance check on an app with no traffic, are named as skipped. It fixes nothing itself. Its folder holds only SKILL.md.

Good for

Convex developers checking an app before launch
Reviewers who want one ranked list instead of four separate audit reports

Install convex-launch-readiness

Claude Code

Copy this prompt.

Install the convex-launch-readiness skill in Claude Code. Run this command:

npx skills add https://github.com/convex/agent-skills --skill convex-launch-readiness --skill convex-deploy-guard --skill convex-authz --skill convex-reviewer --skill convex-advisor --skill convex-insights

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

SKILL.md hands work to convex-deploy-guard, convex-authz, convex-reviewer, convex-advisor and convex-insights, so the first command installs them too. It also mentions convex-migrate-rehearse and convex-suggest as optional hand-offs; install those separately if you want them. Convex's docs say the skills CLI installs skills into .agents/skills/ in your project. The "All Convex skills" command installs every skill in the repo, which has 33 skill folders. One of them, convex-improve-convex-plugin, is listed in the Convex docs as sending your coding session's transcript to the Convex team. Run this skill with /convex-launch-readiness in Claude Code or $convex-launch-readiness in Codex, or let the agent pick it when the task fits. This skill reads your deployment through the official Convex MCP server, which is not part of the skills. Add the server command to your client's MCP settings; the Convex docs also describe an official plugin that starts it for you. This skill works on your project folder, so this page offers no Claude app upload route.

Built from the package data we hold. Not yet tested by us.

What's in the skill

When it loads

Run every Convex audit (authz, reviewer, advisor, insights) into one scored, deduped readiness report with an ordered fix plan — Lighthouse for your backend.

Sections in SKILL.md

Workflow
Rules

Key facts

Author: Convex (GitHub organisation convex; the vendor's own repo)
License: Apache-2.0 (stated in the README and the LICENSE file)
Installs on skills.sh: 38,347 (Nov 2, 2026)
GitHub stars: 68 for the whole convex/agent-skills repo (Nov 2, 2026)
Last repo commit: Oct 29, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
What it includes: 1 file: SKILL.md (the repo tree lists nothing else in the folder)
Part of: The convex/agent-skills repo: 33 skill folders in the tree. The README names 6 skills, and this one is not among them

What it includes

deploy-guard classifies the target, then Claude notes which passes can run and which skip
SKILL.md says a skipped pass is not a pass.
Read · Guard and scope
Runs convex-authz for identity-from-arg, missing ownership, PII leaks and parent writes
Read · Authorization pass
Runs convex-reviewer for validators, indexes, idiom and error handling
Read · Review pass
Runs convex-advisor on live read-limit and conflict evidence, if the app has traffic
Network · Performance pass
Runs convex-insights on recent failures from logs, if a deployment exists
Network · Logs pass
Collapses one defect seen in code and in the deployment into a single finding
The higher-confidence source is kept and nothing is dropped silently.
Normalize and dedupe
Starts at 100, subtracts 15, 5 or 1 per confirmed high, medium or low finding, floor 0
Findings that are only plausible are listed but do not move the score.
Score
Groups blockers, should-fix and nice-to-have, then orders the fixes
Authorization and data-loss issues come first, then scale, then idiom and observability.
Try: “Run a launch-readiness check on this Convex app”
Report and fix plan
Runs the fix skill for each finding you accept, then re-runs passes to show the change
Fix skills named include convex-authz, the reviewer fixers, migrate-rehearse for schema changes and suggest for component swaps.
Runs code · Dispatch fixes on request

From the official docs

Permissions

Medium risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

Accepted fixes edit your project
Dispatch runs fix skills such as convex-authz and migrate-rehearse. SKILL.md says they are gated by their own consent and deploy-target rules. Review each change. github.com
A code-only score is not a production check
When no deployment or traffic exists, the report is labelled code-only. Do not read it as production-verified. github.com
The Convex MCP server limits production access by default
The Convex docs say the MCP server is safe by default: in production deployments agents cannot access PII and can only perform read-only operations. Write access and PII reads on production need --dangerously-enable-production-deployments or --cautiously-allow-production-pii. docs.convex.dev

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	SKILL.md calls the report pass read, aggregate and dispatch. Fixes only run for findings you accept.
Sign-in and keys	SKILL.md names no credentials.
Scopes	Reads your convex/ code and, if a deployment exists, performance and log data through the Convex MCP tools Runs the other audit skills On request, runs fix skills that edit your project

Questions

What does convex-launch-readiness do?

It makes Claude run Convex's audit skills, merge their findings into one report and score the result. You get a ranked list and an ordered fix plan.

How is the score worked out?

It starts at 100 and subtracts 15 for each confirmed high finding, 5 for medium and 1 for low, with a floor of 0. The formula is printed so you can reproduce it. Findings marked only plausible are listed but do not change the number.

Does a high score mean the app is ready for production?

SKILL.md says a green score on a code-only run means "code looks ready", not "production-verified". The report header lists which passes ran and which were skipped.

Does it fix problems?

The skill itself only reads, aggregates and dispatches. When you accept a finding, it runs the matching fix skill, which has its own consent and deploy-target rules, and then re-runs the affected passes.

How do I install it?

Use the install box at the top of this page. The first command installs convex-launch-readiness together with convex-deploy-guard, convex-authz, convex-reviewer, convex-advisor and convex-insights, because the report is built from those passes. The fix skills it dispatches to (such as migrate-rehearse and suggest) are optional. Run it with /convex-launch-readiness in Claude Code or $convex-launch-readiness in Codex.

Is it safe?

The audit passes read your code and, where a deployment exists, usage and log data through the Convex MCP tools. Fix steps edit your project, so review each one. Our advice: run it on dev first and accept fixes one at a time.

Is it official?

It is in the convex/agent-skills repo under the Convex GitHub organisation, which is the vendor's own repo.

Information

updated Nov 2, 2026

Repository: convex/agent-skills
Folder: skills/convex-launch-readiness

convex/agent-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 5

convex/agent-skills README · github.com
skills CLI commands · github.com
Convex docs: Agent Skills · docs.convex.dev
convex-launch-readiness SKILL.md · github.com
Convex docs: MCP server · docs.convex.dev