TokenWatcher AI

Open-source AI observability platform. Monitor AI API costs, tokens, latency & usage in real time. Powered by TokenWatcher.

[TokenWatcher AI @tokenwatcherai]

🚀 Introducing TokenWatcher!
Monitor AI API costs, tokens, latency & usage in real time with an open-source observability platform. Built with TypeScript, React & PostgreSQL.
🎥 Demo: https://t.co/O73hnVL50u
⭐ GitHub: https://t.co/oSDCwH207b

[TokenWatcher AI @tokenwatcherai]

Recent TokenWatcher sprint:
✓ Agent identity
✓ Public agent capabilities
✓ x402 payment flow
✓ Payment verification
✓ Replay protection
✓ Production safety checks
✓ Automated tests
✓ OpenClaw/Telegram integration
Slowly turning a weekend project into infrastructure. ⚡

[TokenWatcher AI @tokenwatcherai]

I added x402 payments to TokenWatcher.
The goal: let agents pay for TokenWatcher capabilities programmatically instead of requiring a human to sit in the middle.
We now have:
→ paid agent capabilities
→ payment verification
→ payment persistence
→ replay protection

[TokenWatcher AI @tokenwatcherai]

One architecture decision I really like in TokenWatcher:
Telegram → OpenClaw → TokenWatcher API → analytics
The agent handles conversation + tools.
Running the OpenClaw layer through ClawUp keeps that separation clean.
Building an agent?
https://t.co/kjrlQHESHE

[TokenWatcher AI @tokenwatcherai]

Building TokenWatcher made me realize I didn't want my AI agent runtime tangled with my core backend.
I run the OpenClaw layer separately and connect it to TokenWatcher's APIs.
ClawUp made that agent workflow much easier to work with.
Try it here: https://t.co/kjrlQHESHE

[TokenWatcher AI @tokenwatcherai]

Just published a new deep dive on TokenWatcher ⚡
The article covers the architecture, intent routing, tools, security, and the actual Telegram flow.
Read it here 👇

https://t.co/3snMjUtM5y

[TokenWatcher AI @tokenwatcherai]

Raw telemetry is useful.
But turning it into answers is where observability gets interesting.
TokenWatcher turns telemetry into:
→ daily cost & requests
→ provider + model spend + latency + errors
→ endpoint costs
→ token usage
All across Overview, Models & Endpoints views.

[TokenWatcher AI @tokenwatcherai]

I wanted TokenWatcher to feel live without constantly polling the backend.
So I built an SSE pipeline, the frontend handles reconnects + stale heartbeats, then refreshes its query state when new telemetry arrives.

No polling loop. Just live telemetry flowing into the dashboard.

[TokenWatcher AI @tokenwatcherai]

The interesting part of TokenWatcher is the telemetry path underneath it.
The SDK:
→ buffers + batches events
→ retries with backoff
→ flushes on shutdown
No “infinite scale” claims.
Just a telemetry pipeline built to avoid turning every event into a network round trip.

[TokenWatcher AI @tokenwatcherai]

The TokenWatcher SDK to keep the developer experience simple while handling the complexity underneath.
track() → bounded queue → batching → retries/backoff → HMAC signing → /ingest
It handles timeouts, aborts and graceful shutdown so telemetry isn't an afterthought. ⚙️

[TokenWatcher AI @tokenwatcherai]

Designing the database behind TokenWatcher.
The requests table acts as the canonical telemetry store, capturing tokens, cost, latency, model, provider, route and metadata.
With workspace-aware relationships and analytics-focused indexes, the same data powers real-time analytics.

[TokenWatcher AI @tokenwatcherai]

What happens when an AI telemetry event reaches TokenWatcher?
The SDK batches and signs the event → /ingest authenticates it → the payload is validated and normalized → telemetry is stored in PostgreSQL → analytics and alerts are updated.⚙️

[TokenWatcher AI @tokenwatcherai]

🏗️ Architecture Spotlight
Every telemetry event follows the same path:
SDK → Backend → Database → Analytics → Dashboard → Telegram
Try it out: https://t.co/qcaYnftbeN
#BuildInPublic #TypeScript #AI

[TokenWatcher AI @tokenwatcherai]

Clean architecture isn't about having more services.
It's about giving every service a clear responsibility.

That's the approach behind TokenWatcher. 🚀
GitHub: https://t.co/oSDCwH207b
ClawUp : https://t.co/kjrlQHESHE

[TokenWatcher AI @tokenwatcherai]

Why doesn't TokenWatcher connect Telegram directly to the backend? 🤔

Because OpenClaw acts as a stateless bridge between Telegram and TokenWatcher.

[TokenWatcher AI @tokenwatcherai]

OpenClaw handles:

• Telegram webhooks
• Intent routing
• Tool execution
• Response rendering
TokenWatcher stays focused on telemetry, analytics and business logic.

[TokenWatcher AI @tokenwatcherai]

Check your AI API usage without opening the dashboard.
TokenWatcher lets you access telemetry directly from Telegram.
📊 Summary
⚡ Recommendations
📈 Forecast
Try it now: https://t.co/qcaYnftbeN

[TokenWatcher AI @tokenwatcherai]

💰 Cost tracking starts at the source.

TokenWatcher records token usage and cost_usd as telemetry is ingested, then builds dashboards and analytics from the same data.

One source of truth. Accurate spend insights. 📊

#AI #TypeScript #OpenSource

[TokenWatcher AI @tokenwatcherai]

A quick look at how TokenWatcher transforms raw telemetry into meaningful insights. Built for observability, scalability, and production-ready AI applications. 📈
Check it out: https://t.co/qcaYnftbeN

[TokenWatcher AI @tokenwatcherai]

💡 Express Tip

TokenWatcher captures the raw request body before parsing JSON, allowing SDK-signed HMAC requests to be verified server-side.

Better security. Replay protection. Workspace-bound telemetry. 🔒

#ExpressJS #TypeScript #OpenSource