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.
npx agentmods add commands/danielsalles/connector-platform-plugin/addgit clone --depth 1 https://github.com/danielsalles/connector-platform-pluginWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00014 | $0.01636 |
| Opus 5 | $0.00007 | $0.00818 |
| Sonnet 5 | $0.00003 | $0.00327 |
| Haiku 4.5 | $0.00001 | $0.00164 |
Grade A, and why
add scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`GET ${CONNECTOR_BASE_URL}/v1/connectors/io.io-platform/<slug>` with bearer. Use Bash + curl. How it starts
The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The user is running /connector add <slug> (e.g. /connector add notion).
Step 1 — Read credentials
Read .env.local. Need CONNECTOR_API_KEY, CONNECTOR_BASE_URL, CONNECTOR_BUILDER_ID. If missing, error:
Not logged in. Run
/connector login <cpt_token>first.
Stop.
Step 2 — Validate the integration exists
GET ${CONNECTOR_BASE_URL}/v1/connectors/io.io-platform/<slug> with bearer. Use Bash + curl.
If 404, reply:
Integration
<slug>not found. Run/connector listto see what's available.
Save the connector's name (human-readable from response).
Step 3 — Detect framework
Read package.json. Required: next in dependencies. If absent, reply:
This plugin currently only scaffolds Next.js (App Router) projects. Open an issue if you need Express, FastAPI, or another framework.
Stop.
Detect App Router vs Pages by checking which exists at the project root:
- If
src/app/orapp/directory exists → App Router (target this) - Else → Pages Router
If Pages Router, reply:
This plugin only scaffolds App Router. Migrate to App Router or open an issue.
Stop.
Determine the source root: src/app if src/ exists, else app.
Step 4 — Write the connect handler
Create <src_root>/api/connect/<slug>/route.ts. If file already exists, ask the user before overwriting.
Use this exact template (substitute <slug> and <name> from step 2):
import { NextResponse } from 'next/server';
// Triggers the connect flow for <name>. Redirects the end_user to the
// authorization page (OAuth or API key form) and returns to the success URL
// after they authorize. The end_user is identified by `external_ref` — replace
// the hardcoded value below with your actual user ID lookup (session, JWT, etc).
//
// Default `__dev__` matches the alias `self` on your MCP URL: connections
// authorized in dev are immediately visible inside Claude Desktop / your MCP
// client. Once you have real customers, swap for your session user ID.
export async function GET(req: Request) {
const url = new URL(req.url);
const externalRef = url.searchParams.get('user_id') ?? '__dev__';
// 1. Ensure the end_user exists in Connector Platform (idempotent).
const userRes = await fetch(`${process.env.CONNECTOR_BASE_URL}/v1/users`, {
method: 'POST',
headers: {
Authorization: `Bearer ${process.env.CONNECTOR_API_KEY}`,
'Content-Type': 'application/json',
},
body: JSON.stringify({ external_ref: externalRef }),
});
if (!userRes.ok) {
return NextResponse.json({ error: 'failed to create end_user', detail: await userRes.text() }, { status: 500 });
}
const { id: endUserId } = await userRes.json();
// 2. Create a connect-session that returns an authorization URL.
const successUrl = `${url.origin}/integrations/<slug>/connected`;
const sessRes = await fetch(`${process.env.CONNECTOR_BASE_URL}/v1/connect-sessions`, {
method: 'POST',
headers: {
Authorization: `Bearer ${process.env.CONNECTOR_API_KEY}`,
'Content-Type': 'application/json',
},
body: JSON.stringify({
end_user_id: endUserId,
connector_namespace: 'io.io-platform',
connector_slug: '<slug>',
success_redirect_url: successUrl,
}),
});
if (!sessRes.ok) {
return NextResponse.json({ error: 'failed to create session', detail: await sessRes.text() }, { status: 500 });
}
const session = await sessRes.json();
return NextResponse.redirect(session.session_url ?? session.url);
}
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.
- 2d ago First seen · 192 lines · 14 tokens per session scan A 86109ce2f4d7
add is a command published in the GitHub repository danielsalles/connector-platform-plugin (0 stars, last pushed 4mo ago), licensed MIT. It adds 14 tokens to every session and 1,636 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.