What Tablif read on Tiltmeter's site
Tiltmeter describes itself as: “Tiltmeter Know when the APIs under your agents move.”
- Stage: Live, no waitlist
- How it differs: “Tiltmeter Learns the contract from real responses, catches undocumented changes and hands what it learns to your agents.”
Where it puts itself
Its one line of difference: “Tiltmeter Learns the contract from real responses, catches undocumented changes and hands what it learns to your agents.”
“OpenAPI diff tools Compare two versions of a spec. Right when the vendor maintains one, blind to what they didn't document. Contract testing Needs the provider to run your contracts, which third-party vendors don't do. Uptime monitors Tell you the API is up. You write and maintain the response assertions yourself. Tiltmeter Learns the contract from real responses, catches undocumented changes and hands what it learns to your agents.”
What it costs
No free plan.
- Free plan: no
“Open source · MIT · Python 3.12+”
Who it sells to
Sells to developers. In Python.
- Sells to: developers
- Languages: Python
“An MCP server lets Claude Code, Cursor, LangChain or any MCP client call has_drifted before trusting an API, and get_schema to see which fields to expect.”
“Open source · MIT · Python 3.12+”
How it runs
A command-line tool. Runs locally and works offline. Distributed through the npm. Open source. AI is a side feature.
- Kind: cli
- Runs where: on the device, offline
- Distributed via: npm
- Open source: yes
- AI role: a side feature
“$ pip install tiltmeter copy”
“One SQLite file, no account. Shapes, status codes, latencies and changes. Response bodies only if you turn samples on.”
What it works with
Names Anthropic, Groq and 5 more on its pages.
- Names: Anthropic · Groq · HubSpot · +4
“Anthropic. Swapping the model id isn't enough: Claude Sonnet 5.5, the recommended replacement, rejects temperature and thinking budgets.”
“Groq. 260 of 1,619 popular, active AI repositories still name a model Groq has retired”
What it offers as proof
Live, no waitlist. Shows customer logos. Claims “1.29M”. Publishes docs.
- Customer logos: yes
- Number claimed: 1.29M
“Groq. 260 of 1,619 popular, active AI repositories still name a model Groq has retired”
“GitHub stars across the repositories that got a fix 1.29M”
Sources
Tablif read these pages, last on 27 Oct 2026. Everything above is what the product says about itself; Tablif does not verify every claim.
Similar products in Tablif Town
- Meter Gateway: See what your AI coding agents spend and touch
- Agentmetry: A local-first endpoint sensor for AI coding agents
- Codemind: Delegate coding. Get back code that works.
- AgentMingo: from ticket to pull request with coding agents
- Tuning Forks: Building AI agents for everyone
- Probe: Native API client for humans and agents.
- APIFreaks: API hub for developers with a wide range of high-quality data APIs and more.
- Hallmonitor: See and answer every coding agent from your notch