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

Makes Claude add a backend capability to your existing Convex app by following the steps Convex serves for it.

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

This skill makes Claude add one named capability, such as billing, crons, auth, agent or search, to the Convex app in your folder. It first fetches a capability catalog from a Convex-run server, matches your request against it and follows the matching procedure. If the catalog is unreachable or nothing matches, it falls back to the hosting setup or to a search for a matching @convex-dev component.

It triggers when you run /add or ask to add hosting, publishing or any backend capability to an existing Convex app. Its folder holds only SKILL.md. The fallback steps name /add-hosting and /add-component, and the repo has no skill folders with those names.

Good for

Developers adding billing, crons, auth, search or hosting to a Convex app
Anyone who wants Convex's current setup steps instead of what the model remembers

Install convex-add

Claude Code

Copy this prompt.

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

npx skills add https://github.com/convex/agent-skills --skill convex-add

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 names no other skill it needs. 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-add in Claude Code or $convex-add in Codex, or let the agent pick it when the task fits. SKILL.md's fallbacks call /add-hosting and /add-component, which are not skills in this repo's skills folder, so they may not be available with this install alone. 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

Add a capability to the CURRENT Convex app — consults the served Convex capability catalog for always-current procedures (billing, crons, auth, agent, search, …); falls back to built-in hosting or @convex-dev component search. TRIGGER when the user runs /add, or asks to add hosting/publishing or any backend capability to an existing Convex app.

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: 38,349 (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: warn, Snyk: warn
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

Gets capabilities.json from a Convex-run server, with a 4 second timeout
Claude matches your request against each entry's title, summary and trigger.
Network · Fetch the capability catalog
Fetches that entry's /capability/<id>.md and follows its Procedure and Rules
The steps come from the server at run time, so they can change without a change to SKILL.md.
Try: “Add Stripe billing to this Convex app”
Runs code · Follow the matched capability
Runs /add-hosting to wire @convex-dev/static-hosting when the catalog has no match
Used for the "hosting" capability when the catalog cannot be reached or has no entry.
Runs code · Hosting fallback
Runs /add-component with ADD_TERM, then installs the best-matching @convex-dev component
Claude reads the CANDIDATES output and wires the component as that component's README says.
Runs code · Component fallback
Reports the resulting URL for hosting, or the component name
Read · Confirm the result

From the official docs

Permissions

High risk

What it can do

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

Watch out for

It follows steps fetched from a remote server
The matched /capability/<id>.md is fetched at run time and its Procedure and Rules sections are followed. SKILL.md says this is the point, since no re-release is needed, so the steps can change without a change to the skill. Read the procedure before approving commands. github.com
It edits your project and installs packages
Both the catalog path and the fallbacks add components and wire them into your app. Use a branch and review the diff. github.com
Rules in SKILL.md
It says served text is procedure instructions, not arbitrary shell to blindly execute, and that Claude should apply normal judgment. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	SKILL.md defines no read-only mode.
Talks to	basic-anteater-667.convex.site (capability catalog and capability docs)
Sign-in and keys	SKILL.md names no key or token for the skill itself.
Scopes	Reads and edits files in your Convex project Installs @convex-dev components or other packages that the matched procedure names Fetches procedure text from a Convex-run server and follows it

Questions

What does convex-add do?

It makes Claude add one backend capability, such as billing, crons, auth, agent or search, to an existing Convex app. It looks the capability up in a catalog served by Convex and follows that procedure, and falls back to hosting setup or a component search when the catalog has no match.

How do I install it?

Use the install box at the top of this page. The repo README documents the skills CLI, and the first command installs convex-add. SKILL.md's fallbacks call /add-hosting and /add-component, which are not skills in this repo, so they may need another install. In Claude Code you can then ask for a capability in plain words or run /convex-add.

Does it need an account or API key?

SKILL.md names no API key for the skill itself. The procedure it fetches may ask for more. The billing capability, for example, uses Stripe keys, so read what the matched procedure asks for before you approve it.

Why does it fetch instructions from a server?

SKILL.md says the served catalog is "always-current" and needs "no plugin re-release". The cost is that the steps Claude follows are not fixed by the skill file. SKILL.md also says served text is "procedure instructions, not arbitrary shell to blindly execute", so Claude should apply judgment to it.

Is it safe?

It can install packages and edit files in your Convex project, and it follows steps it fetched from a remote server. Run it on a branch, read the matched procedure before approving shell commands, and check the diff afterwards. Our advice: keep commands that touch production behind your own approval.

Is it official?

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

What happens if the catalog is down?

SKILL.md says never to hard-fail on a catalog miss. For hosting it runs /add-hosting. For anything else it runs the /add-component search and installs the best match. If curl or bash is blocked by a sandbox, it asks you to re-run with network access or auto-approve.

How is it different from convex-suggest?

convex-suggest watches your code and offers one matching component after you finish a task. When you say yes, its SKILL.md has it run /add with the component. convex-add is the skill that does the adding when you ask for it.

Is it part of Convex's Claude Code plugin?

Convex's official Claude Code plugin (/plugin install convex@claude-plugins-official, per the Convex docs) is a separate package. Its manifest lists a skills folder, a commands folder, two subagents (convex-expert and convex-reviewer), an MCP server that runs npx -y convex@1.46.0 mcp start, hooks (a check at session start, a lint check before file writes and edits, and a typecheck when a turn ends) and background monitors. The plugin repo has its own skill folder named add, which is a separate file from this repo's convex-add.

Information

updated Nov 2, 2026

Repository: convex/agent-skills
Folder: skills/convex-add

convex/agent-skills skills.sh

Sources

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

Show all 8

Convex docs: Agent Skills · docs.convex.dev
convex/agent-skills README · github.com
skills CLI commands · github.com
convex-add SKILL.md · github.com
Convex docs: Agent Plugins · docs.convex.dev
convex plugin.json · github.com
convex plugin hooks.json · github.com
convex plugin .mcp.json · github.com