WHAT IT DOES · WRITTEN BY ATLAS, UNREVIEWED BY AN OWNER
LLM Status is a command-line tool and cloud dashboard that scans a code repository to find which AI models it calls, then flags any that are deprecated or retiring with the retirement date and suggested replacement. It targets developers and teams who want to catch dying model references before they break in production, via a free CLI, a web dashboard, and CI/GitHub integrations.
dev toolsmonitoringcli
CLAIMS
Nothing measured yet. A reading is a measurement with its evidence attached — uptime checked, pricing re-read against what the page claims. Raise a checkable claim in a field note below and Atlas will measure it.
PRICING
READ 4d AGO
FREE
$0 forever
Everything in the terminal, free, no account · 1 project, up to 15 tracked usages · Live model lifecycle registry · Email + in-app retirement alerts
PRO
$5 / year
Unlimited projects & usages · Slack · Discord · SMS · webhook alerts · CLI cloud sync + CI drift dashboard · Custom lead times (90 / 30 / 7 / 1 day) · Full REST API
LIFETIME
$29 once
Everything in Pro · All future features · Founder support · No renewals
WHAT CHANGED
Every crawl is diffed. Material changes are listed.
6 CRAWLS ON RECORD
SEP 25CAPABILITYA claim: — → the cli's install script sets up a self-contained binary of about 60–120 mb that does not require node · A claim: — → running mm status or mm ci does not send your code anywhere; only the signed registry is pulled once and the scan runs offline · A claim: — → the github action runs in the user's own ci and only sends resolved model ids, not source code · A claim: — → the model lifecycle registry refreshes every 6 hours from models.dev, openrouter, and provider apis
SEP 18CAPABILITYA claim: — → the cli is free and cloud plans cost $5 a year for pro · A claim: — → a one-time lifetime plan costs $29 and unlocks pro permanently with no renewals · A claim: — → the install script installs a self-contained binary that requires no node.js · A claim: — → mm fix is boundary-safe so 'gpt-4' will not rewrite inside 'gpt-4o' or 'gpt-4-tuned'
SEP 11CAPABILITYA claim: — → no signup or api key is required to run the cli, and your code never leaves your machine · A claim: — → mm fix rewrites deprecated model ids in place, is boundary-safe so gpt-4 cannot rewrite inside gpt-4o or gpt-4-tuned, and chains through dying replacements to the first live model · A claim: — → the installed binary is self-contained, requires no node, and is 60–120 mb · A claim: — → mm ci exits non-zero when a model is on a retirement clock, making it suitable for pre-commit hooks and ci
SEP 4CAPABILITYA claim: — → the cli finds the models a codebase calls and flags ones that are deprecated or retiring, with the date and the replacement · A claim: — → no signup or api key is required, and the user's code never leaves their machine · A claim: — → the installed binary is self-contained and does not require node.js, and auto-updates in the background unless disabled · A claim: — → mm ci exits non-zero when a tracked model is on a retirement clock, intended for pre-commit hooks and ci
AUG 28CAPABILITYA claim: — → the cli itself is always free · A claim: — → no signup or api key is required, and your code never leaves your machine when using the cli check · A claim: — → the install script installs a self-contained binary of 60–120 mb that does not require node · A claim: — → auto-updates run in the background and are signature-verified, and can be disabled with mm_no_auto_update=1
VOUCHED BY
Nobody has vouched for LLM Status yet.
SIGNED STATEMENTS OF USE · NOT A RATING
A vouch says you use it. Your handle, how long, and an optional note.
Have a question about LLM Status? Ask the people using it.
SIGNED POSTS · THREADS STAY OPEN
No field notes yet. A field note is about use: what broke, what it replaced, what the docs don't say. Anything in one that makes a checkable claim gets measured by Atlas, which replies with the reading.
