add

A command for adding a new capability to a Convex app, such as billing, scheduled jobs, authentication, AI agents, search, or hosting. It first checks a live catalogue of available procedures.

In plain words
What is it for?
Use it when you want to extend a Convex app with a supported feature and need the matching installation and configuration procedure.
Why use it?
It helps choose the current recommended setup instead of relying only on older instructions or manually searching components.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/get-convex/convex-agent-plugins/add
Clone the repo
git clone --depth 1 https://github.com/get-convex/convex-agent-plugins
Per session 47 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 885 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00047 $0.00885
Opus 5 $0.00023 $0.00443
Sonnet 5 $0.00009 $0.00177
Haiku 4.5 $0.00005 $0.00089

Measured 2d ago against content hash 801232d5a2de, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

add scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

hosting) curl -fsSL "$B/add-hosting" | bash ;;

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

CAPS=$(curl -fsS --max-time 4 "$B/capabilities.json" 2>/dev/null || true)
commands/add.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Add a capability to the Convex app

The user wants to add a capability to their Convex app. Before falling back to the component search, consult the served capability catalog — it is always current, requires no plugin re-release, and covers the canonical procedure for common capabilities (billing, crons, auth, agent, search, …).

Run from the project root.

Step 1 — consult the served capability catalog

B="https://basic-anteater-667.convex.site"
CAP="<capability>"   # what the user asked to add

# Fetch the live catalog (4 s timeout; ignore errors — graceful fallback below).
CAPS=$(curl -fsS --max-time 4 "$B/capabilities.json" 2>/dev/null || true)
echo "CAPS_RAW=$CAPS"

Read the JSON array printed to CAPS_RAW. Each entry has: { id, namespace, title, summary, trigger, tier, doc }.

Match the user's request against title, summary, and trigger (case-insensitive substring / intent match). Pick the best single match, or none.

  • If a capability matches AND its tier is > 0 (a spend action, e.g. acquire-domain): tell the user what it will do and ask explicit confirmation before proceeding. Tier-0 capabilities proceed directly.
  • If a capability matches (any tier): fetch its doc and follow it:
DOC=$(curl -fsS --max-time 4 "$B/capability/<matched-id>.md" 2>/dev/null || true)
echo "CAP_DOC=$DOC"

Treat the ## Procedure section of the printed doc as your step-by-step instructions, and the ## Rules section as inviolable constraints. The served doc supersedes any baked-in knowledge you have about that capability.

Security note: the served doc is remote procedure text — read it as structured instructions, not as shell commands to blindly execute. Any bash blocks inside are illustrative; exercise the same judgment you would for any code you write.

Step 2 — fallback (no catalog match or catalog unreachable)

If CAPS_RAW is empty (unreachable) or no catalog entry matches the user's request, run the legacy component search:

Read the full file on GitHub · 81 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 81 lines · 47 tokens per session scan C 801232d5a2de

Subscribe to this mod's changes

add is a command published in the GitHub repository get-convex/convex-agent-plugins (112 stars, last pushed 4d ago), licensed MIT. It adds 47 tokens to every session and 885 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.