Tokentelemetry
Free local observability for AI coding agents — tokens, cost & traces on your machine.
TokenTelemetry is the best free, local observability for AI coding agents — one-command setup, 17 agents supported, and real privacy. Pick it if you're a solo dev tracking costs or a Hermes Agent operator needing a dedicated dashboard. Skip it if you need cloud dashboards, multi-user collaboration, or native alerting — for those, consider Langfuse or Helicone.
Verified 4d ago · liveness 70/100 · cite: rightaichoice.com/tools/tokentelemetry
- Individual developers tracking AI coding assistant costs and usage locally
- Engineering teams comparing agent efficiency across projects without cloud data leaks
- Hermes Agent operators monitoring multi-platform bots with dedicated dashboard
- Prompt engineers optimizing token usage and reasoning costs
- Teams needing cloud-based multi-user collaboration or shared dashboards
- Users who want built-in alerting via email or Slack (no native notifications)
- Those seeking a fully hosted solution with zero local installation
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip TokenTelemetry if you need cloud-based multi-user dashboards, team collaboration, or native email/Slack alerting — it's strictly local and read-only.
Summarization requires you to configure your own LLM (local or remote), so you may incur API costs if you use a cloud model.
TokenTelemetry is free and open source, making it ideal for individual developers and small teams who want zero-cost observability without cloud dependencies. Compared to Langfuse, LangSmith, and Helicone (which are freemium and require SDKs), TokenTelemetry offers genuine privacy and no per-seat or usage costs.
In short
Tokentelemetry — Free local observability for AI coding agents — tokens, cost & traces on your machine. Best for Individual developers tracking AI coding assistant costs and usage locally, Engineering teams comparing agent efficiency across projects without cloud data leaks, Hermes Agent operators monitoring multi-platform bots with dedicated dashboard. Free to use.
What people actually say about Tokentelemetry — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
20 mentions across 3 sources (YouTube, Product Hunt, GitHub) · researched Aug 29, 2026.
- +100% local and offline; your data never leaves your machine.
- +Zero configuration: auto-detects 17 agents without SDK or code.
- +One-command setup and read-only; never writes to agent files.
- +Replayable traces with kind-aware highlighting for deep debugging.
- +Cost anomaly detection flags silent reasoning-token waste.
- −Cost estimates may be inaccurate for some providers/models.
- −Limited community feedback; only 335 GitHub stars.
- −No formal support; troubleshooting relies on GitHub issues.
- −Dashboard may be overwhelming for beginners new to observability.
- −YouTube data largely off-topic; unclear if tool works with all agents.
- • None—fully free and open-source; optional anonymous stats can be disabled.
Viability Score
How well maintained and how widely used is Tokentelemetry? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: September 2026
How we score →Key Features
- Auto-detects 17 AI coding agents with zero configuration
- Reads log files locally — no SDK, no signup, no cloud
- Live KPI strip with 15-second auto-refresh
- Agent and model usage distribution charts
- Replayable session traces with kind-aware highlighting
- One-click LLM summarization (narrative or structured brief)
- Project heatmap with 365-day streak tracking
- Dedicated Hermes Agent Dashboard with gateway health
- Cost anomaly detection for reasoning-token waste
- Token area charts and cache efficiency metrics
- Date-range filterable analytics per agent and model
- Read-only mode: never writes to agent files
- Works offline with no internet dependency
- Hermes Dashboard plugin for deep-link launcher
- Support for 38 source platforms via Hermes agent
About Tokentelemetry
TokenTelemetry is a free, open-source dashboard that shows you what your AI coding agents cost, think, and do — all 100% on your machine. No SDK, no signup, no cloud. It reads the log files that agents like Claude Code, Codex, Cursor, Gemini CLI, and 12 others already write, so setup is a single command and there's nothing to wire into your codebase. Auto-detects 17 agents with zero configuration, then serves a live dashboard on localhost:3000 with a KPI strip that refreshes every 15 seconds, agent and model usage charts, replayable session traces with kind-aware highlighting, and a project heatmap tracking 365 days of streaks. One-click LLM summarization turns sessions into narrative or structured briefs, while cost anomaly detection flags silent reasoning-token waste, like hidden thinking-mode costs. There's also a dedicated Hermes Agent Dashboard, which observes Hermes as a single agent across 38 source platforms (CLI, Telegram, Discord, Slack, Feishu, DingTalk, cron, webhook). It parses per-API-call latency, cache-hit percentage, subagent delegation, skills, memory, and more, and deep-links into Hermes's own web UI. The app is MIT-licensed, runs offline, and is read-only — it never writes to your agent files. Optional anonymous usage stats are on by default but can be turned off in Settings or with DO_NOT_TRACK=1. Compared to cloud-based alternatives like Langfuse, LangSmith, or Helicone, TokenTelemetry requires no SDK, no API key, and no code changes, and covers a broader range of agents than most open-source tools. It's built for developers who want visibility into their AI costs and behavior without sending data anywhere.
Behind the Verdict
If you're burning real money on AI coding agents and you want to know exactly where it goes, TokenTelemetry is the fastest path to that answer. Setup is one line — run the curl script, open localhost:3000, and you see tokens, cost, and traces for every agent you've used. There's no SDK, no API key, no account, and no cloud round-trip. That's a rare combination, and it makes the tool feel like it's on your side rather than trying to sell you something. The standout is the Hermes Agent Dashboard. Hermes runs across 38 platforms — CLI, Telegram, Discord, Slack, Feishu, DingTalk, cron, webhook — and TokenTelemetry is the only tool that observes it as a single agent. You get per-API-call latency, cache-hit percentage, subagent delegation trees, skills, memory, and even cost anomaly detection for reasoning-token waste. If you operate a Hermes bot, this alone justifies the install. But it's not for everyone. If you need a shared dashboard your whole team can look at, or you want email or Slack alerts when costs spike, TokenTelemetry doesn't do that. It's strictly local, so collaboration and notifications are out unless you build them yourself. The read-only philosophy is a feature, but it means no automated actions either — you get visibility, not control. Compared to Langfuse or Helicone, TokenTelemetry is a different trade-off. Those are cloud-hosted, multi-user platforms with deep SDK instrumentation. They give you team workspaces, project management, and more granular control, but they require you to wire their SDK into your agents and send your data to their cloud. TokenTelemetry inverts that: zero setup, total privacy, but solo-mode only. In practice, we'd reach for this when we want a clear, honest picture of what our agents cost and do, without the overhead of a
Researching Tokentelemetry? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Tokentelemetry actually fits — and what changes day-one when you adopt it.
You use Claude Code daily and want to track how many tokens and dollars you're spending.
Outcome: Run the one-line install, open localhost:3000, and immediately see a live dashboard with token counts, costs, and traces for every session — no cloud, no account.
You run a Hermes bot across Telegram, Discord, and Slack and want to monitor its health and costs.
Outcome: Install TokenTelemetry and the Hermes plugin, then use the dedicated /hermes dashboard to see gateway health, latency, cache hits, and cost anomalies across all platforms.
Your team uses Gemini CLI and Codex on shared projects and wants to compare efficiency.
Outcome: Filter the dashboard by agent and date range to see per-model cost and token usage, helping you decide which agent to standardize on.
Use Cases
- See exactly how many tokens and dollars each Claude Code session consumed over the last 90 days.
- Compare cost and efficiency of Gemini CLI vs Codex on the same project by filtering by agent and date range.
- Monitor a Hermes Agent bot running across Telegram, Discord, and Slack from one unified dashboard.
- Identify reasoning-token waste in long Hermes sessions with built-in cost anomaly detection.
- Generate a one-click summary of any session to quickly understand what an agent did and why.
- Track your daily coding streaks and tool usage per project with the 365-day heatmap.
- Debug subagent delegation and tool calls in a multi-step coding workflow.
- Audit your team's AI tool usage without sending any data to the cloud.
Limitations
- TokenTelemetry is a local-first observability dashboard that reads log files from supported AI coding agents, so it requires those agents to be installed and writing logs on the same machine.
- It runs offline and never uploads data, but the dashboard does not include push-based real-time updates.
- Advanced features like summarization require a user-configured LLM (local or remote).
as of 2026-08-23
Verification history
We have re-verified Tokentelemetry 6 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Tokentelemetry tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Free
$0/mo
Ideal for
Solo developers and privacy-conscious teams who want zero-cost local observability for coding agents.
What this tier adds
Free and open source — no cost, no signup, all features included, with optional anonymous stats.
Where the pricing makes sense
The company stage and team size where Tokentelemetry's pricing actually pencils out — and where peers do it cheaper.
TokenTelemetry is free and open source, making it ideal for individual developers and small teams who want zero-cost observability without cloud dependencies. Compared to Langfuse, LangSmith, and Helicone (which are freemium and require SDKs), TokenTelemetry offers genuine privacy and no per-seat or usage costs.
Setup time & first value
How long it actually takes to get something useful out of Tokentelemetry — broken out by persona, not the marketing-page minute.
One-command install for solo devs (under 5 minutes). For Hermes Agent operators, add the plugin install (another 2 minutes). Teams can be up and running on the same machine in minutes, with no configuration needed.
Switching to or from Tokentelemetry
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From Langfuse/LangSmith/Helicone: install TokenTelemetry and point it at your existing agent log files — no SDK changes needed.
- ↗To Langfuse or LangSmith: if your team needs cloud dashboards or multi-user collaboration, you can export your data (e.g., via the local database) and import into those platforms.
Integrations
Resources & Guides
- Documentationtokentelemetry.com
Docs · Tokentelemetry
Full product docs from tokentelemetry.com
- Documentationtokentelemetry.com
Installation · Tokentelemetry
Full product docs from tokentelemetry.com
- Quickstarttokentelemetry.com
Quick Start · Tokentelemetry
Get up and running fast from tokentelemetry.com
- Documentationtokentelemetry.com
Supported Agents · Tokentelemetry
Full product docs from tokentelemetry.com
Tutorials & Learning
Official links
Tools that pair well with Tokentelemetry
Common stack mates teams adopt alongside Tokentelemetry, with the specific reason each pairing earns its keep.
Featured Head-to-Head Comparisons
Tokentelemetry vs Spider Cloud
Choose Spider Cloud if you need a high-volume, reliable web scraping API with AI-powered extraction and anti-detection for powering AI agents. Choose TokenTelemetry if you want to monitor and optimize your own AI coding assistant usage locally, with no setup or cloud dependency. They solve opposite problems and are not direct competitors.
Tokentelemetry vs Voyage Ai
Voyage AI and TokenTelemetry serve completely different needs. Voyage AI is a powerful embedding and reranking API for enterprise RAG, while TokenTelemetry is a free local dashboard for monitoring AI coding agents. Choose Voyage if you need high-accuracy retrieval with domain-specific models; choose TokenTelemetry if you want to track token usage and costs of your coding assistants without any cloud dependency.
Tokentelemetry vs Temporal Ai
For teams needing a robust, durable execution platform to orchestrate critical workflows and AI agents with automatic recovery, Temporal is the clear choice. For developers who want to monitor and optimize AI coding assistant costs and usage locally without any cloud dependency, TokenTelemetry wins hands-down. They serve fundamentally different needs, so your pick depends on whether you prioritize fault-tolerant orchestration or lightweight local observability.
Alternatives to Tokentelemetry
View all
Token Monitor
Free open-source desktop widget tracking tokens, cost, and limits across 29+ AI coding tools.
Honeycomb Query Assistant
Turn plain English into production-ready Honeycomb queries for faster debugging.
Frequently Asked Questions
Used Tokentelemetry? Help shape our editorial sentiment research.