A
license-
qualityD
maintenanceHuman-in-the-Loop authorization gateway for AI Agents. Securely pause MCP workflows and route high-risk actions to human approvers via Slack or Email.
Last updated
51
1
MIT
Impri
by sekera-radim
Impri provides a human-approval inbox for AI agents, enabling human oversight of agent actions and automated monitoring of external sources.
Action Approval Workflow
Submit actions for approval (
impri_push_action): Push a proposed action (e.g., sending an email, posting a comment) with a title, formatted preview, optional context URL, and editable fields so reviewers can tweak content before approving. Supports idempotent submissions to prevent duplicates on retry.Await a human decision (
impri_await_decision): Poll until the human approves, rejects, or the action expires — returns the final decision along with any edits made by the reviewer.Report execution result (
impri_report_result): After attempting an approved action, report whether it succeeded or failed, closing the audit loop.Check inbox queue depth (
impri_inbox_status): See how many actions are pending to avoid flooding the inbox and causing approvals to expire.
Watcher / Monitoring
Create a custom watcher (
impri_create_watcher): Set up recurring monitors on external sources (RSS feeds, Reddit, URL diffs, GitHub releases, etc.) with keyword filters, deduplication, and configurable schedules — delivering matches to the inbox or a webhook.List active watchers (
impri_list_watchers): View all configured watchers, their status (active, paused, degraded), and IDs.Discover watcher presets (
impri_list_watcher_presets): Browse 18+ pre-built templates for common sources like Hacker News, Reddit, GitHub releases, npm, YouTube, arXiv, and more.Create a watcher from a preset (
impri_create_watcher_from_preset): Instantly spin up a watcher by choosing a preset and supplying minimal parameters (e.g., a subreddit name or GitHub repo).
Additional Capabilities
Notification channels: Integrate with Slack, Discord, Telegram, or email for pending approval alerts.
Audit log: Maintains a full trail of events, decisions, and execution results for transparency.
CLI: A dedicated command-line interface for managing the inbox, approving/rejecting actions, and managing watchers and API keys.
Self-hosting: Deploy via Docker Compose or a local dev setup, with a full REST API and OpenAPI spec.
Provides a watcher preset to monitor arXiv for new papers and request human approval before taking action.
Integrates with CrewAI to incorporate approval steps into multi-agent workflows.
Sends notifications to Discord channels for pending actions requiring approval.
Provides watcher presets to monitor GitHub releases and events, enabling approval-gated actions.
Integrates with LangChain to add human approval gating in AI agent workflows.
Integrates with Make to add human approval steps in automated scenarios.
Integrates with n8n to enable approval-based workflows in automation pipelines.
Provides a watcher preset to monitor npm package releases and request approval for related actions.
Sends push notifications via ntfy for pending actions, enabling quick human review.
Integrates with OpenAI Agents to require human approval before executing agent actions.
Provides watcher presets to monitor Reddit posts and comments, with approval workflows for responses.
Sends notifications to Slack channels and allows approval decisions via Slack integration.
Sends notifications and provides in-chat Approve/Reject buttons via a Telegram bot for fast approvals.
Integrates with Zapier to connect approval workflows with hundreds of other apps.
Impri — Approval Inbox for AI Agents
The imprimatur for your AI agents. Watchers watch the world, the Approval Inbox holds the agent's hands until a human says yes.
Quickstart
Two ways to run Impri — pick one. Both give you an API key and an inbox URL in under 5 minutes.
Cloud (no install)
curl -s -X POST https://api.impri.dev/v1/signup \
-H "Content-Type: application/json" \
-d '{"name": "my-agent"}'
# → { "key": "im_...", "project_id": "proj_...", ... }Or skip curl and click Create an API key at app.impri.dev — same result. Your inbox is at app.impri.dev, the API base URL is https://api.impri.dev/v1. It's early beta but is the fastest way to try Impri with no Docker required.
Docker Compose (self-host, < 5 minutes)
git clone https://gitlab.com/sekera.radim/impri.git
cd impri
docker compose upOpen http://localhost:8080 in your browser.
On first start the server prints the bootstrap Admin API key to the logs:
╔══════════════════════════════════════════════════════╗
║ IMPRI — FIRST RUN BOOTSTRAP ║
╠══════════════════════════════════════════════════════╣
║ Admin API Key: im_... ║
║ Project ID: proj_... ║
║ Store this key securely — it will not be shown again.║
╚══════════════════════════════════════════════════════╝Copy the key, paste it into the login screen, and you're in.
Dev mode (hot-reload, self-host)
Terminal 1 — server:
cd server
npm install
npm run dev
# Server starts on http://localhost:8484Terminal 2 — UI:
cd ui
npm install
npm run dev
# UI starts on http://localhost:5173
# /v1 requests are proxied to localhost:8484Related MCP server: final-approval
API at a glance
Base URL: https://api.impri.dev/v1 (cloud) or http://localhost:8484/v1 (self-host)
Auth: Authorization: Bearer im_<key>
Method | Path | Description |
POST |
| Push a new action for approval |
GET |
| List actions ( |
GET |
| Get action detail + decision |
POST |
| Approve or reject (single) |
POST |
| Approve or reject up to 50 actions at once |
POST |
| Report execution result |
GET |
| OpenAPI spec |
Push an action (curl example)
curl -X POST https://api.impri.dev/v1/actions \
-H "Authorization: Bearer im_..." \
-H "Content-Type: application/json" \
-d '{
"kind": "reddit.comment",
"title": "Reply to: Why is resume advice so conflicting?",
"preview": {
"format": "markdown",
"body": "The advice conflicts because..."
},
"target_url": "https://reddit.com/r/jobs/comments/...",
"expires_in": 86400,
"editable": ["preview.body"]
}'Self-hosting instead? Swap the URL for http://localhost:8484/v1/actions.
MCP server (Claude Code / agents)
npx @impri/mcp
# cloud: IMPRI_API_KEY=im_... IMPRI_BASE_URL=https://api.impri.dev
# self-host: IMPRI_API_KEY=im_... IMPRI_BASE_URL=http://localhost:8484Project structure
server/ TypeScript + Fastify + SQLite — REST API (port 8484)
mcp/ MCP server (stdio) — thin wrapper over the REST API
ui/ Vue 3 + Vuetify — web inbox (port 5173 dev / 8080 Docker)
docker/ Dockerfiles (server.Dockerfile)
docs/ Research, ADRsCLI
The impri CLI lets humans manage the inbox from a terminal — approve, reject, tail pending actions, add watchers, and manage keys — without writing any code.
# Build and install (local, pre-npm)
cd sdk/typescript && npm install && npm run build
cd ../cli && npm install && npm run build
npm install -g ./cli
# Connect to your instance
impri init --cloud --signup # or: impri init (self-hosted)
# Common commands
impri inbox # pending actions
impri tail # live-tail new actions
impri approve act_abc123
impri watch add github-releases --param owner=fastify --param repo=fastifySDKs & integrations
v0.1, pre-release — both cloud and self-host work today; expect rough edges either way.
Package | Location | Language |
CLI |
| Node 18+ |
Python SDK |
| Python 3.10+ |
TypeScript SDK |
| Node 18+ (native fetch) |
MCP server |
| Any MCP client |
pip install -e sdk/python # Python SDK (local, pre-PyPI)
npm install ./sdk/typescript # TS SDK (local, pre-npm)
npx @impri/mcp # MCP server (published)Integrations — LangChain, OpenAI Agents, CrewAI, n8n, Make, Zapier, webhook receivers
Cookbook — recipes for email approval, SQL gating, social posts, idempotent batches, webhook verification, key rotation
Documentation
Web docs: https://impri.dev/docs
CLI reference — install,
impri init, every command with examples, config + env precedenceQuickstart — signup → first approved action in < 5 min
Example agent — a complete, dependency-free agent that proposes an action, waits for approval, then acts and reports back
Self-hosting — Docker, env vars, backups, reverse proxy
Webhooks — HMAC verification, retries, polling fallback
Inbox UX & Bulk API — keyboard shortcuts, bulk approve/reject, search/filter parameters,
POST /v1/actions/bulk-decisionreferenceWatcher presets — 18 ready-to-use templates (HN, Reddit, GitHub, npm, arXiv, …); REST + SDK + MCP usage
Notification channels — Slack, Discord, Telegram, ntfy, email, and generic webhook; digest window, auto-disable, SSRF protection
Telegram Approval Bot — in-chat Approve / Reject buttons; setup, security model, troubleshooting
Audit log — event types, query API (
GET /v1/audit), export (NDJSON/CSV), retention, and security modelllms.txt— machine-readable index for AI assistants
Self-hosting notes
SQLite data is persisted in a Docker volume (
impri-data).Set
WEBHOOK_SECRETenv var to a random string for HMAC webhook signing.BASE_URLshould match the public URL of your deployment (used in inbox_url links).
License
MIT — see LICENSE. Self-host the full core freely; the hosted cloud
and team features are the paid offering (see MONETIZATION.md).
Maintenance
–Maintainers
–Response time
–Release cycle
–Releases (12mo)
Commit activity
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseAqualityDmaintenanceHuman-in-the-loop approval gate for AI agents. Your agent calls submit_approval before any irreversible action; a human reviews on a branded page; a signed webhook fires back with the decision.Last updated11MIT
- Alicense-qualityCmaintenanceProvides a human approval gate for AI agents, enabling interactive inline cards for approving, editing, or rejecting actions before they are executed.Last updatedMIT
- Alicense-qualityAmaintenanceLocal-first AI agent for approval-gated automation and verifiable LLM workflows.Last updated1MIT
Related MCP Connectors
Human-in-the-loop for AI coding agents — ask questions, get approvals via Slack.
Authenticated email gateway for AI agents — per-agent inboxes, HITL approval, SPF/DKIM verified.
Authenticated email gateway for AI agents — per-agent inboxes, HITL approval, SPF/DKIM verified.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/sekera-radim/impri'
If you have feedback or need assistance with the MCP directory API, please join our Discord server