Individual
Free
1 person
- Personal AI adoption dashboard
- Claude Code, Codex, and Cursor sync
- Desktop MCP connector
- Email support
You’re paying for AI.
Is it paying off?
Adoption, cost, and impact across your AI tools — in one secure view.
Works with the tools your team already uses
OpenAI
Claude
Gemini
xAI
Mistral
DeepSeek
AWS Bedrock
Azure OpenAIADOPTION & VISIBILITY
Prove adoption, cost, and impact
No more guessing, and no more piecing it together across a dozen dashboards. See who’s using AI, what it’s costing, and what you’re getting back.
Team Adoption
Last 30 days
Adoption Rate
87%+12%
Active Engineers
34/39+5
Avg Sessions
6.2/day+18%
Team Cost
$3,847+12%
Adoption Heatmap
See daily adoption patterns across your entire team
Per-Engineer View
Who's active, who needs training, and who's leading
Model Preferences
Which models your team reaches for most
Adoption Consistency
Track adoption trends week over week
PRIVATE TEAM BENCHMARKS
Team benchmarks that stay private.
See how teams across your org stack up — who’s thriving, who’s stalled, and where to step in. Private to your org, visible only to the people you choose.
Activity
Who is showing the strongest AI adoption right now
Consistency
Who keeps adoption steady day after day
Cache Efficiency
Who is creating the strongest cost impact
Adoption
Who is using AI tools across the broadest surface area
Private and secure by default
Scoreboards are scoped to your organization and protected by role-based access control. Only owners, admins, and authorized members can view rankings. Workspace member names stay inside this private internal view. The public leaderboard uses aliases only. It never uses real names, account names, or emails.
Trust & Privacy
Security-first by design
Whether you connect via Admin API keys, wrapped proxy keys, or MCP telemetry — your credentials are encrypted, we do not store prompt or code content, and you control what gets shared.
Connect local tools
One command authenticates your machine and installs the CostHawk MCP for Claude Code, Codex, Cursor, and Gemini CLI.
npm exec --yes costhawk@latest -- --login
CompatibleClaude CodeOpenAI Codex CLICursorGemini CLI
AES-256 key encryption
Admin API keys are encrypted at rest with AES-256-GCM. We never store or see your keys in plaintext.
Wrapped key isolation
Proxy your API calls through CostHawk. Your real provider keys never touch your codebase or client devices.
Local-first parsing
MCP telemetry is computed locally from supported developer-tool data directories before anything leaves your machine.
No prompt storage
We only persist usage metadata: token counts, models, timestamps, and hashed project IDs. Prompt and code content is never stored.
Preview before upload
Dry-run syncs show exactly what would be sent, with payload previews and optional file lists for full transparency.
Opt-in auto-sync
Automatic uploads are off by default. Enable with COSTHAWK_AUTO_SYNC=true and disable Codex sync anytime.
Ask CostHawk
Give leaders a live AI cost analyst.
CostHawk's MCP turns your AI assistant into an authorized reporting surface for usage, savings, anomaly detection, pricing, alert, and ROI questions. CTOs get answers in plain language; the dashboard remains the system of record.
Want the full tool map?
The CostHawk MCP server exposes 22 tools for usage reporting, local sync, savings analysis, anomaly detection, alerts, pricing, proxy setup, ROI, and workspace guardrails.
npm exec --yes costhawk@latest -- --login
Claude CodeCodexCursorGemini CLI
Leadership questions
“What changed in AI usage since last week?”
“Which providers or projects are driving spend?”
“Are flat-rate subscriptions saving us money?”
“What anomalies need attention before month-end?”
“Create a Slack budget alert for this workspace.”
Weekly operating rhythm
Ask for usage, cost, and model movement before the next leadership meeting.
Subscription decisions
Compare retail token cost against Claude, ChatGPT, and team plans.
Exception management
Surface spikes, inspect alerts, and route budget issues into Slack or PagerDuty.
Rollout signal
Use local-tool metadata to see whether adoption is real or just announced.
USE CASES
The operating view for AI adoption.
Engineering leaders and CTOs get a tailored view of adoption, cost, and impact across their team. Useful for product and finance leaders too.
Engineering Lead
See who's actually using AI
Spot champions, find stalled teams, turn habits into standards.
CTO
See where adoption is stuck
One operating view across teams, tools, and providers.
CPO
See where AI is shipping product
Connect AI investment to real roadmap throughput.
CFO
See AI cost before it sprawls
Tracked cost, impact, and budget risk in one view.
FEATURES
Operational controls for adoption, cost, and impact.
Set budgets, route API calls through wrapped proxy keys, sync provider data automatically, and export finance-ready reports — all from one place.
Budget Alerts
Get notified before costs exceed your limits
Cost Optimizer
Find cheaper models without sacrificing quality
Savings Engine
Compare subscription vs. pay-per-use plans
Data Export
CSV, JSON, or formatted reports for invoicing
Admin API Connections
Connect provider admin keys, sync automatically
Ask CostHawk
Authorized AI assistants can answer usage, savings, and anomaly questions
Wrapped Proxy Keys
Route API calls for attribution and cost limits
5-Minute Setup
Full visibility with no infrastructure changes
REST API
Access adoption and cost data programmatically
WHY COSTHAWK
The shortcut to AI visibility.
Everything engineering, product, and finance leaders need to answer the board — without building an internal reporting system from scratch.
Get the answer before the board asks
Know exactly who on your team is using AI, how often, and whether adoption is growing or stalling.
See the full picture in one place
Every provider, every engineer, every project — unified in a single dashboard instead of scattered across billing portals.
Built for action, not just reporting
Budget alerts, cost optimization recommendations, and savings analysis so you can act on what you see.
No procurement headache
Start free as an individual, then add Team when you need teammates in the workspace.
Pricing
Simple pricing. No surprises.
Individual is free for one person. Start a Team trial when you need teammates.
Team
$99/mo
2-100 people
30-day free trial
- Everything in Individual
- Team adoption roster
- Project and team membership
- Owner/admin org reporting
Enterprise
Custom
101+ people
- Everything in Team
- Custom integrations
- SSO / SCIM
- Dedicated support + SLA
For organizations with 101+ people or custom integrations / support.
Individual is free for one person. Team includes a 30-day trial; add a card when you upgrade or before day 31.
FAQ
Questions, answered.
What is CostHawk?+
CostHawk is an AI adoption monitoring platform for engineering teams. It tracks who on your team is using AI tools, how much they cost, and how adoption is trending — across OpenAI, Anthropic, Google, and other providers.
How does CostHawk track AI adoption across my team?+
CostHawk supports three integration methods: MCP telemetry for local AI-tool activity and assistant-accessible reports, admin API sync for organization-wide provider cost data, or wrapped proxy keys that sit between your code and the provider.
What can CostHawk's MCP server do for leadership teams?+
CostHawk's MCP server lets authorized leaders and operators ask their AI assistant for usage summaries, savings analysis, cost anomalies, live model pricing, budget alerts, webhooks, local sync, and ROI reports. It is the conversational layer on top of the same CostHawk data in the dashboard.
How are team benchmarks kept private?+
Yes. Team Benchmarks are scoped to your organization and protected by role-based access control. Workspace member names are visible only to authorized members. The public leaderboard uses aliases only. It never uses real names, account names, or emails.
What providers does CostHawk support?+
OpenAI, Anthropic, Google, xAI, Mistral, Cohere, DeepSeek, Groq, Meta, Perplexity, Together, AWS Bedrock, and Azure OpenAI. We add new providers regularly.
What data does CostHawk see — and not see?+
CostHawk stores telemetry metadata: token counts, model names, and timestamps. We never see, store, or process your prompts, responses, or code.
How do I get started?+
Sign up, invite your team, and connect via MCP telemetry or admin API keys. The entire setup takes under 5 minutes.