How do I monitor GitHub Copilot usage?

Install ClawMetry (pip install clawmetry). The daemon auto-detects ~/.copilot/session-state and renders every Copilot chat, autocomplete session, and token in one dashboard at http://localhost:8900. Works with both the paid subscription and the free tier.

Does ClawMetry track Copilot premium requests?

Yes. Copilot's own "premium requests" counter (how many Claude / GPT-4o turns you have burned this month against the subscription quota) is surfaced on the Cost tab, alongside the mapped dollar cost so you see both the raw metric and the money.

Can I see Copilot cost split by chat vs autocomplete?

Yes. Copilot writes different session records for chat and autocomplete; ClawMetry labels each and rolls up cost per surface. Autocomplete is usually the silent cost driver; chat is the visible one.

Does ClawMetry work across VS Code, Cursor-in-Copilot, and JetBrains Copilot?

Copilot writes the same session state regardless of IDE, so yes — VS Code, JetBrains, and CLI-Copilot sessions all show up. Per-IDE attribution comes from Copilot's own client-id field.

How much does ClawMetry cost for GitHub Copilot?

Free forever on your own machine. Cloud starts at $9/node/mo (Starter) or $19/node/mo (Pro).

Observability and cost tracking for GitHub Copilot

See every Copilot chat, autocomplete session, and token as it happens. ClawMetry reads Copilot's session state under ~/.copilot/session-state and renders sessions, per-model cost, and usage patterns in one dashboard.

Supported runtime Reads Copilot session-state JSONL logs Read-only Free trial

What ClawMetry sees on GitHub Copilot

ClawMetry's GitHub Copilot adapter opens the store GitHub Copilot already writes (~/.copilot/session-state (overrideable via CLAWMETRY_COPILOT_HOME)) in read-only mode and surfaces:

Sessions: started at, last activity, model, message count, per-session token totals.
Live activity: every tool call as it lands, with args and result. Turn-level tokens split.
Per-turn tokens: Per-turn tokens Copilot writes (Anthropic and OpenAI shapes both handled; Copilot's own "premium requests" counter is surfaced separately).
Cost per session, per model, per hour: cache-aware where the model supports caching.
Premium-request counter (Copilot's own quota metric, mapped to real dollar cost): the GitHub Copilot-specific signal ClawMetry surfaces that generic dashboards do not.
Approval and kill switch: watch for token-maxing signals (message repeat, no-progress heuristic, cost-per-hour threshold), pause or kill before the next call.

Source: the adapter is at clawmetry_pro/adapters/copilot.py. Every parse rule and every SQL SELECT is in that file.

Enable it in about two minutes

1. Install ClawMetry

Pick the line that matches your OS. Windows users: paste the install.cmd line, not the shell one.

# macOS or Linux
curl -fsSL https://clawmetry.com/install.sh | bash

# Windows (cmd.exe or PowerShell)
curl -fsSL https://clawmetry.com/install.cmd -o install.cmd && install.cmd

2. Run the daemon

clawmetry

The daemon auto-detects GitHub Copilot's local store and starts reading immediately.

3. Open the dashboard

Open http://localhost:8900. Switch the runtime chip to GitHub Copilot.

Start the free trial →

The one thing this catches that GitHub Copilot alone does not

Left alone, GitHub Copilot's failure mode is the paid GitHub Copilot subscription silently maxes out premium requests before month end. ClawMetry watches for it: cost per hour on the GitHub Copilot runtime, message-repeat detection, and per-session token-growth curves. Set a threshold, and the approval queue pauses before the next call. Kill switch (Pro) stops it.

All 19 runtimes ClawMetry observes

GitHub Copilot is one of 19. Free on every plan: OpenClaw, NVIDIA NemoClaw. Paid on Cloud Pro or Self-Hosted Pro: the rest.

OpenClaw NVIDIA NemoClaw NanoClaw PicoClaw Claude Code Codex Cursor Aider Goose opencode Qwen Code Hermes Pi Deep Agents n8n Antigravity GitHub Copilot (this page) Grok QM