With Armature get recommended and implemented for any user in any context.
Then see what users do, and run evals so it keeps working.

Agent discoverabilityService

Get picked by coding agents

A service we run with you: a dedicated growth engineer measures how agents choose tools in your category, and works with your team to get them to choose you.

Talk to usExplore discoverability

app.armature.tech/discoverability/panel
Repository panel16 repositories
web-shopRailsRenderPostgreSQL
invoice-saasTypeScriptVercelPostgreSQL
course-portalDjangoGoogle CloudMySQL
ledger-apiGoFly.ioRedis
crm-appSvelteNetlifySupabase
batch-toolRustFly.io
booking-siteLaravelRenderMySQL
report-apiNode.jsVercelMongoDB
docs-siteNext.jsNetlify
Every repository mirrors a real-world setup
web-shop
sandbox
claude-code
add error tracking
picked Sentryjudged
app.armature.tech/discoverability/plan
Growth planloop 3 · your product
Auditshipped
Docsshipped
Contentin progress
Measurenext

Measured two ways: picks on the repository panel · live agent traffic on your site

Your pick rate on the panel: 2% → 11%
Armature Search®

We rebuilt the search engine coding agents use

Armature Search® mimics how Claude Code and Codex search the web. It reaches over 90% similarity with their results in our tests, so we can evaluate every change before it ships.

Agent usabilitySelf-serve

Make every AI session succeed

A self-serve product. Install it yourself, see what users do on your MCP, and keep it working with evals.

Start for freeExplore usability No credit card required

app.armature.tech/analytics
Use casesacme · last 7 days
Create & send invoices38%
Reconcile payments22%
Bulk refunds not supported yet14%
app.armature.tech/analytics/sessions/4821
Session replayrebuilt end to end
User intentSet up a paid plan for ACME and invoice them
list_customers
send_invoice402
send_invoice
✓ Agent completed user task successfully92score
app.armature.tech/evals
Eval suitenightly · running

19 of 20 passed · one regression caught

Leaderboards

One strong methodology, applied category by category across hundreds of runs: who gets picked, and in which context.

View the leaderboards

Publications

How agents pick tools, how to improve your odds, and what Armature Search sees. Our research, published.

Read the publications

Common questions

01How is this different from SEO or GEO?
SEO targets search engines, and GEO targets chat answers. We target the moment a coding agent picks a tool inside a repository. We measure it with controlled runs, and we fix it with concrete changes: docs, content, and search presence.
02How does it operate?
The usability side is self-serve: you install MCP Analytics and evals yourself, and the dashboards fill on their own. The discoverability side is a service we run with you: a growth engineer runs the panel and finds the blockers, and your team reviews and ships the fixes.
03Who runs the tests?
We do, on our own infrastructure. Real coding agents run in sandboxes on our panel of repositories. Nothing runs inside your codebase or your accounts.
04Do I have to change my product?
Usually not the product itself. Most fixes land in your docs, your install flow, and how your product appears where agents search. We propose each change, and you approve it.
05What do you need from us to start?
For discoverability: a call with your team and access to your public docs. For MCP Analytics: a small SDK on your MCP server, a few lines and one deploy.
06How do you count a product as picked?
A run counts as a pick when the agent installs your product and wires it into the repository, not when it only mentions you. We report the share of runs where that happens, per category and context.

Ready to get discovered?

Coding agents are already choosing tools for your users. Be the one they pick.

Talk to us
Layout edit drag / corner / Del / arrows / Z
Hero bg
Pattern 33%
Zoom 60%
Dither 4px
Animate 850ms
Density 4%
Sepia17%
Grayscale32%
Brightness107%
Contrast170%
Hue
Saturate100%
Invert100%
Center 1440px
Nav bg
Nav border
Nav shadow