AnnouncementsFrom the team
Research
September 2, 2026
Which tools do coding agents install? What 3,547 judged runs across 18 sectors show
We ran Claude Code and Codex inside 51 realistic repositories, on 481 frozen prompts, across 18 developer tool sectors. A judge read every session. How we built the experiment, what the agents chose, and where the method is weak.
Read the article →
August 4, 2026
Launching Armature Evals - Agent Evaluations for MCPs and CLIs
Real agents replay your users' workflows, and a judge scores every run. Catch regressions before you ship.
July 22, 2026
Launching Armature - Product Analytics for MCPs, Claude Connectors and ChatGPT Apps
See what users do on your MCP: sessions rebuilt end to end, grouped into use cases and issues.
Coming soon
What we built with Armature Search®
The search engine we made to mimic how coding agents search the web, and what it lets us test.