# The Armature library > Guides, comparisons and data on how coding agents such as Claude Code, Codex and Cursor choose which software to install. Source: https://armature.tech/library Publisher: Armature, Inc. Every page below is also available as Markdown: add `.md` to the address. ## Guides Long pieces that explain a whole subject from the start. - [Agent discoverability: the complete guide](https://armature.tech/library/agent-discoverability) — Coding agents now choose which software gets installed. Agent discoverability is the work of getting chosen: what it is, and what moves it. - [AI visibility for developer tools](https://armature.tech/library/ai-visibility-for-developer-tools) — AI visibility dashboards count brand mentions in chat answers. For a developer tool the number that pays rent is install share. - [Answer engine optimization, explained](https://armature.tech/library/answer-engine-optimization) — AEO is the work of being the answer rather than a link. What it is, what actually moves it, and where it stops working for a developer tool. - [Developer tool marketing in the agent era](https://armature.tech/library/developer-tool-marketing-in-the-agent-era) — The developer marketing playbook was built for a human funnel. A large share of tool decisions now happen inside a coding agent. - [Documentation for coding agents](https://armature.tech/library/documentation-for-coding-agents) — Your documentation is now read mostly by machines that paste your code and check whether it runs. What to change, in the order that pays. - [Evals for an MCP server or a CLI](https://armature.tech/library/evals-for-mcp-servers-and-clis) — Unit tests check your code. Evals check whether a real agent can still do the job through it. What to test, and why a judge is needed. - [Generative engine optimization for developer tools](https://armature.tech/library/generative-engine-optimization-for-developer-tools) — GEO was built for brands that want mentioning in chat answers. Developer tools have a second, larger surface: the coding agent. - [How coding agents choose tools](https://armature.tech/library/how-coding-agents-choose-tools) — What happens between a developer's one-line request and a package landing in the lock file. The steps, and what changes the outcome. - [How we measured this](https://armature.tech/library/how-we-measured-this) — The method behind every number here: 51 realistic codebases, 481 frozen prompts, real agents at pinned versions, and a blind judge. - [MCP analytics: seeing what users do through an agent](https://armature.tech/library/mcp-analytics) — When a user delegates to an agent, the session happens inside Claude or ChatGPT and your interface never sees it. What to capture. - [Technical content marketing when the reader is a machine](https://armature.tech/library/technical-content-marketing-for-ai-search) — Agents search with long, site-scoped queries and want configuration, limits and pricing units. Most content answers something else. - [What about Gemini CLI, Windsurf, Cline and the others?](https://armature.tech/library/what-about-the-other-coding-agents) — We measured three agents and not the rest. Why those three, what we think transfers to the others, and what we are not willing to claim without data. ## Comparisons How agent discoverability differs from the tools next to it. - [AEO vs GEO vs SEO, and the fourth one nobody names](https://armature.tech/library/aeo-vs-geo-vs-seo) — Three acronyms for overlapping work, and a fourth surface that none of them measure. A plain comparison with the differences that actually change what you do. - [Agent discoverability vs generative engine optimization](https://armature.tech/library/agent-discoverability-vs-geo) — Both are about being chosen by AI. One ends in a sentence a person reads, the other in a package in a lock file. What separates them. - [Armature vs Ahrefs Brand Radar](https://armature.tech/library/armature-vs-ahrefs-brand-radar) — Brand Radar adds AI visibility to an SEO suite you may already pay for. Armature measures whether coding agents install your product. Where each fits. - [Agent discoverability vs hiring a developer advocate](https://armature.tech/library/armature-vs-hiring-a-devrel) — A DevRel hire costs more than the service and does different work. What each one is actually for, and which one to do first. - [Armature vs LangSmith, Langfuse and Braintrust](https://armature.tech/library/armature-vs-llm-observability-tools) — LLM observability watches the agents you build. Armature watches the agents your users bring. Same word, opposite direction, and most teams need both. - [Armature vs Otterly](https://armature.tech/library/armature-vs-otterly) — Otterly is the cheapest way to see what AI answers say about you. Armature measures whether coding agents install your product. - [Armature vs Peec AI](https://armature.tech/library/armature-vs-peec-ai) — Peec AI tracks brand visibility across AI assistants for a low monthly price. Armature measures whether coding agents install you. - [Armature vs Profound](https://armature.tech/library/armature-vs-profound) — Profound measures how AI assistants describe your brand to people. Armature measures whether coding agents install your product. - [Armature vs Scrunch AI](https://armature.tech/library/armature-vs-scrunch-ai) — Scrunch AI is a full monitoring and optimisation suite for AI search. Armature measures whether coding agents install your product. - [Armature vs the Semrush AI Toolkit](https://armature.tech/library/armature-vs-semrush-ai-toolkit) — The Semrush AI Toolkit tracks AI visibility inside a suite you may already have. Armature measures installs by coding agents. What each one is for. - [The best AI visibility tools for developer tools in 2026](https://armature.tech/library/best-ai-visibility-tools-for-developer-tools) — A buyer's guide to GEO and AEO platforms, what each one measures, what they cost, and the one surface none of them cover if you sell developer tools. - [A developer marketing agency vs agent discoverability](https://armature.tech/library/developer-marketing-agency-vs-agent-discoverability) — Draft.dev, Hackmamba and Infrasity produce good technical content. None can tell you whether Claude Code installed your package. ## From the experiment Pages built from the 5,292 judged runs. - [Do Claude Code, Codex and Cursor pick the same tools?](https://armature.tech/library/do-claude-code-and-codex-agree) — All three agents agreed on the category leader in only 8 of 18 categories. Where they differ, why, and what it means if you measure only one. - [Do coding agents recommend Algolia?](https://armature.tech/library/do-coding-agents-recommend-algolia) — Algolia was chosen in 5% of 459 judged search sessions, ranking sixth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend ALTCHA?](https://armature.tech/library/do-coding-agents-recommend-altcha) — ALTCHA was chosen in 8% of 160 judged bot protection sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Amazon CloudWatch?](https://armature.tech/library/do-coding-agents-recommend-amazon-cloudwatch) — Amazon CloudWatch was chosen in 8% of 360 judged observability sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Amazon RDS for PostgreSQL?](https://armature.tech/library/do-coding-agents-recommend-amazon-rds-for-postgresql) — Amazon RDS for PostgreSQL was chosen in 5% of 356 judged databases sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Amazon S3?](https://armature.tech/library/do-coding-agents-recommend-amazon-s3) — Amazon S3 is the most chosen file storage service, taking 46% of 90 judged file storage sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Amazon SES?](https://armature.tech/library/do-coding-agents-recommend-amazon-ses) — Amazon SES was chosen in 12% of 208 judged email sending sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Amplitude?](https://armature.tech/library/do-coding-agents-recommend-amplitude) — Amplitude was chosen in 4% of 359 judged product analytics sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Anthropic SDK?](https://armature.tech/library/do-coding-agents-recommend-anthropic-sdk) — Anthropic SDK was chosen in 1% of 681 judged agent frameworks sessions, ranking number 14. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Arize Phoenix?](https://armature.tech/library/do-coding-agents-recommend-arize-phoenix) — Arize Phoenix was chosen in 8% of 288 judged evals sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Atlas Search?](https://armature.tech/library/do-coding-agents-recommend-atlas-search) — Atlas Search was chosen in 4% of 459 judged search sessions, ranking eighth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Auth0?](https://armature.tech/library/do-coding-agents-recommend-auth0) — Auth0 was chosen in 23% of 201 judged authentication sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Autocannon?](https://armature.tech/library/do-coding-agents-recommend-autocannon) — Autocannon was chosen in 6% of 216 judged performance testing sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend AWS Lambda?](https://armature.tech/library/do-coding-agents-recommend-aws-lambda) — AWS Lambda is the most chosen serverless platform, taking 24% of 287 judged serverless compute sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend AWS X-Ray?](https://armature.tech/library/do-coding-agents-recommend-aws-x-ray) — AWS X-Ray was chosen in 3% of 360 judged observability sessions, ranking ninth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend AWS?](https://armature.tech/library/do-coding-agents-recommend-aws) — AWS is the most chosen cloud provider, taking 62% of 215 judged cloud sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure AI Foundry Agent Service?](https://armature.tech/library/do-coding-agents-recommend-azure-ai-foundry-agent-service) — Azure AI Foundry Agent Service was chosen in 3% of 681 judged agent frameworks sessions, ranking eighth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure Blob Storage?](https://armature.tech/library/do-coding-agents-recommend-azure-blob-storage) — Azure Blob Storage was chosen in 20% of 90 judged file storage sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure Communication Services Email?](https://armature.tech/library/do-coding-agents-recommend-azure-communication-services-email) — Azure Communication Services Email was chosen in 14% of 208 judged email sending sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure Database for PostgreSQL Flexible Server?](https://armature.tech/library/do-coding-agents-recommend-azure-database-for-postgresql-flexible-server) — Azure Database for PostgreSQL Flexible Server was chosen in 8% of 356 judged databases sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure Functions?](https://armature.tech/library/do-coding-agents-recommend-azure-functions) — Azure Functions was chosen in 13% of 287 judged serverless compute sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure SQL Database?](https://armature.tech/library/do-coding-agents-recommend-azure-sql-database) — Azure SQL Database was chosen in 10% of 356 judged databases sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Azure Voice Live API?](https://armature.tech/library/do-coding-agents-recommend-azure-voice-live-api) — Azure Voice Live API was chosen in 3% of 308 judged voice agents sessions, ranking seventh. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Better Auth?](https://armature.tech/library/do-coding-agents-recommend-better-auth) — Better Auth was chosen in 4% of 201 judged authentication sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Better Stack?](https://armature.tech/library/do-coding-agents-recommend-better-stack) — Better Stack was chosen in 4% of 360 judged observability sessions, ranking sixth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend BigQuery?](https://armature.tech/library/do-coding-agents-recommend-bigquery) — Coding agents raised BigQuery in 66 of 356 judged databases sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Braintree?](https://armature.tech/library/do-coding-agents-recommend-braintree) — Coding agents raised Braintree in 64 of 395 judged payments sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Braintrust?](https://armature.tech/library/do-coding-agents-recommend-braintrust) — Braintrust was chosen in 6% of 288 judged evals sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Checkly?](https://armature.tech/library/do-coding-agents-recommend-checkly) — Checkly was chosen in 5% of 360 judged observability sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Claude Agent SDK?](https://armature.tech/library/do-coding-agents-recommend-claude-agent-sdk) — Claude Agent SDK was chosen in 2% of 681 judged agent frameworks sessions, ranking ninth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Claude Managed Agents?](https://armature.tech/library/do-coding-agents-recommend-claude-managed-agents) — Claude Managed Agents was chosen in 2% of 681 judged agent frameworks sessions, ranking number 11. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Clerk?](https://armature.tech/library/do-coding-agents-recommend-clerk) — Clerk was chosen in 10% of 201 judged authentication sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Cloudflare AI Gateway?](https://armature.tech/library/do-coding-agents-recommend-cloudflare-ai-gateway) — Cloudflare AI Gateway was chosen in 21% of 140 judged AI gateways sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Cloudflare R2?](https://armature.tech/library/do-coding-agents-recommend-cloudflare-r2) — Cloudflare R2 was chosen in 13% of 90 judged file storage sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Cloudflare Turnstile?](https://armature.tech/library/do-coding-agents-recommend-cloudflare-turnstile) — Cloudflare Turnstile is the most chosen bot protection service, taking 57% of 160 judged bot protection sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Cloudflare Workers?](https://armature.tech/library/do-coding-agents-recommend-cloudflare-workers) — Cloudflare Workers was chosen in 7% of 287 judged serverless compute sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Cloudflare?](https://armature.tech/library/do-coding-agents-recommend-cloudflare) — Cloudflare was chosen in 11% of 270 judged deployment sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend CodeSandbox SDK?](https://armature.tech/library/do-coding-agents-recommend-codesandbox-sdk) — Coding agents raised CodeSandbox SDK in 70 of 299 judged code sandboxes sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Cursor SDK?](https://armature.tech/library/do-coding-agents-recommend-cursor-sdk) — Cursor SDK was chosen in 14% of 681 judged agent frameworks sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Datadog?](https://armature.tech/library/do-coding-agents-recommend-datadog) — Datadog was chosen in 4% of 360 judged observability sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Daytona?](https://armature.tech/library/do-coding-agents-recommend-daytona) — Daytona was chosen in 16% of 299 judged code sandboxes sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend DBOS Transact?](https://armature.tech/library/do-coding-agents-recommend-dbos-transact) — DBOS Transact was chosen in 2% of 681 judged agent frameworks sessions, ranking tenth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend django-axes?](https://armature.tech/library/do-coding-agents-recommend-django-axes) — django-axes was chosen in 7% of 160 judged bot protection sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Docker?](https://armature.tech/library/do-coding-agents-recommend-docker) — Docker came up in 36% of judged code sandboxes sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend DuckDB?](https://armature.tech/library/do-coding-agents-recommend-duckdb) — Coding agents raised DuckDB in 68 of 356 judged databases sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend E2B?](https://armature.tech/library/do-coding-agents-recommend-e2b) — E2B is the most chosen code sandbox, taking 42% of 299 judged code sandboxes sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Elasticsearch?](https://armature.tech/library/do-coding-agents-recommend-elasticsearch) — Elasticsearch was chosen in 5% of 459 judged search sessions, ranking seventh. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend ElevenLabs Agents?](https://armature.tech/library/do-coding-agents-recommend-elevenlabs-agents) — ElevenLabs Agents was chosen in 10% of 308 judged voice agents sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Firecracker?](https://armature.tech/library/do-coding-agents-recommend-firecracker) — Firecracker came up in 80% of judged code sandboxes sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend GitHub Pages?](https://armature.tech/library/do-coding-agents-recommend-github-pages) — GitHub Pages was chosen in 6% of 270 judged deployment sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend GlitchTip?](https://armature.tech/library/do-coding-agents-recommend-glitchtip) — GlitchTip was chosen in 3% of 360 judged observability sessions, ranking tenth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend GoCardless?](https://armature.tech/library/do-coding-agents-recommend-gocardless) — GoCardless was chosen in 2% of 395 judged payments sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Google Analytics?](https://armature.tech/library/do-coding-agents-recommend-google-analytics) — Coding agents raised Google Analytics in 91 of 359 judged product analytics sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Google Cloud Run?](https://armature.tech/library/do-coding-agents-recommend-google-cloud-run) — Google Cloud Run was chosen in 4% of 287 judged serverless compute sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Google Cloud Storage?](https://armature.tech/library/do-coding-agents-recommend-google-cloud-storage) — Google Cloud Storage was chosen in 20% of 90 judged file storage sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Google Cloud?](https://armature.tech/library/do-coding-agents-recommend-google-cloud) — Google Cloud was chosen in 17% of 215 judged cloud sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Grafana k6?](https://armature.tech/library/do-coding-agents-recommend-grafana-k6) — Grafana k6 was chosen in 4% of 216 judged performance testing sessions, ranking sixth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Grafana?](https://armature.tech/library/do-coding-agents-recommend-grafana) — Grafana was chosen in 14% of 360 judged observability sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend gVisor?](https://armature.tech/library/do-coding-agents-recommend-gvisor) — gVisor came up in 69% of judged code sandboxes sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend hCaptcha?](https://armature.tech/library/do-coding-agents-recommend-hcaptcha) — Coding agents raised hCaptcha in 126 of 160 judged bot protection sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Honeycomb?](https://armature.tech/library/do-coding-agents-recommend-honeycomb) — Honeycomb was chosen in 4% of 360 judged observability sessions, ranking eighth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend hyperfine?](https://armature.tech/library/do-coding-agents-recommend-hyperfine) — hyperfine was chosen in 6% of 216 judged performance testing sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Inngest?](https://armature.tech/library/do-coding-agents-recommend-inngest) — Inngest was chosen in 7% of 681 judged agent frameworks sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Inspect AI?](https://armature.tech/library/do-coding-agents-recommend-inspect-ai) — Inspect AI was chosen in 4% of 288 judged evals sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend JMH?](https://armature.tech/library/do-coding-agents-recommend-jmh) — JMH is the most chosen performance testing tool, taking 10% of 216 judged performance testing sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend JWT?](https://armature.tech/library/do-coding-agents-recommend-jwt) — JWT came up in 61% of judged authentication sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend Langfuse?](https://armature.tech/library/do-coding-agents-recommend-langfuse) — Langfuse is the most chosen eval tool, taking 34% of 288 judged evals sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend LangGraph?](https://armature.tech/library/do-coding-agents-recommend-langgraph) — LangGraph was chosen in 3% of 681 judged agent frameworks sessions, ranking seventh. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend LangSmith?](https://armature.tech/library/do-coding-agents-recommend-langsmith) — LangSmith was chosen in 3% of 288 judged evals sessions, ranking sixth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Laravel AI SDK?](https://armature.tech/library/do-coding-agents-recommend-laravel-ai-sdk) — Laravel AI SDK was chosen in 1% of 681 judged agent frameworks sessions, ranking number 13. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Lemon Squeezy?](https://armature.tech/library/do-coding-agents-recommend-lemon-squeezy) — Coding agents raised Lemon Squeezy in 94 of 395 judged payments sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend LiteLLM?](https://armature.tech/library/do-coding-agents-recommend-litellm) — LiteLLM was chosen in 18% of 140 judged AI gateways sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend LiveKit Agents?](https://armature.tech/library/do-coding-agents-recommend-livekit-agents) — LiveKit Agents was chosen in 11% of 308 judged voice agents sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend LlamaIndex?](https://armature.tech/library/do-coding-agents-recommend-llamaindex) — Coding agents raised LlamaIndex in 64 of 681 judged agent frameworks sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Mailgun?](https://armature.tech/library/do-coding-agents-recommend-mailgun) — Coding agents raised Mailgun in 97 of 208 judged email sending sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Meilisearch?](https://armature.tech/library/do-coding-agents-recommend-meilisearch) — Meilisearch was chosen in 9% of 459 judged search sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Microsoft Azure?](https://armature.tech/library/do-coding-agents-recommend-microsoft-azure) — Coding agents raised Microsoft Azure in 76 of 215 judged cloud sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Microsoft Entra ID?](https://armature.tech/library/do-coding-agents-recommend-microsoft-entra-id) — Microsoft Entra ID was chosen in 10% of 201 judged authentication sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend MinIO?](https://armature.tech/library/do-coding-agents-recommend-minio) — Coding agents raised MinIO in 87 of 215 judged cloud sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Modal?](https://armature.tech/library/do-coding-agents-recommend-modal) — Modal was chosen in 25% of 299 judged code sandboxes sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Mollie?](https://armature.tech/library/do-coding-agents-recommend-mollie) — Mollie was chosen in 3% of 395 judged payments sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend MongoDB Text Search?](https://armature.tech/library/do-coding-agents-recommend-mongodb-text-search) — MongoDB Text Search was chosen in 2% of 459 judged search sessions, ranking number 11. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend MySQL FULLTEXT?](https://armature.tech/library/do-coding-agents-recommend-mysql-fulltext) — MySQL FULLTEXT was chosen in 2% of 459 judged search sessions, ranking tenth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Neon?](https://armature.tech/library/do-coding-agents-recommend-neon) — Neon is the most chosen database, taking 66% of 356 judged databases sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend New Relic?](https://armature.tech/library/do-coding-agents-recommend-new-relic) — New Relic was chosen in 4% of 360 judged observability sessions, ranking seventh. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenAI Agents SDK?](https://armature.tech/library/do-coding-agents-recommend-openai-agents-sdk) — OpenAI Agents SDK was chosen in 4% of 681 judged agent frameworks sessions, ranking sixth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenAI Assistants?](https://armature.tech/library/do-coding-agents-recommend-openai-assistants) — Coding agents raised OpenAI Assistants in 70 of 681 judged agent frameworks sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenAI Evals?](https://armature.tech/library/do-coding-agents-recommend-openai-evals) — Coding agents raised OpenAI Evals in 72 of 288 judged evals sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenAI Realtime API?](https://armature.tech/library/do-coding-agents-recommend-openai-realtime-api) — OpenAI Realtime API was chosen in 12% of 308 judged voice agents sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenRouter?](https://armature.tech/library/do-coding-agents-recommend-openrouter) — OpenRouter was chosen in 11% of 140 judged AI gateways sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenSearch?](https://armature.tech/library/do-coding-agents-recommend-opensearch) — OpenSearch was chosen in 13% of 459 judged search sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend OpenTelemetry?](https://armature.tech/library/do-coding-agents-recommend-opentelemetry) — OpenTelemetry came up in 68% of judged observability sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend Oracle Indexed Search?](https://armature.tech/library/do-coding-agents-recommend-oracle-indexed-search) — Oracle Indexed Search was chosen in 5% of 459 judged search sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Paddle?](https://armature.tech/library/do-coding-agents-recommend-paddle) — Paddle was chosen in 4% of 395 judged payments sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend PayPal?](https://armature.tech/library/do-coding-agents-recommend-paypal) — Coding agents raised PayPal in 139 of 395 judged payments sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend PHPBench?](https://armature.tech/library/do-coding-agents-recommend-phpbench) — PHPBench was chosen in 6% of 216 judged performance testing sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend PlanetScale?](https://armature.tech/library/do-coding-agents-recommend-planetscale) — Coding agents raised PlanetScale in 63 of 356 judged databases sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Portkey?](https://armature.tech/library/do-coding-agents-recommend-portkey) — Portkey is the most chosen AI gateway, taking 21% of 140 judged AI gateways sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Postgres Full-Text Search?](https://armature.tech/library/do-coding-agents-recommend-postgres-full-text-search) — Postgres Full-Text Search is the most chosen search engine, taking 17% of 459 judged search sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend PostgreSQL?](https://armature.tech/library/do-coding-agents-recommend-postgresql) — PostgreSQL came up in 95% of judged databases sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend PostHog?](https://armature.tech/library/do-coding-agents-recommend-posthog) — PostHog is the most chosen product analytics tool, taking 53% of 359 judged product analytics sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Postmark?](https://armature.tech/library/do-coding-agents-recommend-postmark) — Postmark was chosen in 27% of 208 judged email sending sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Prism?](https://armature.tech/library/do-coding-agents-recommend-prism) — Prism was chosen in 4% of 681 judged agent frameworks sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Prometheus?](https://armature.tech/library/do-coding-agents-recommend-prometheus) — Coding agents raised Prometheus in 139 of 360 judged observability sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Promptfoo?](https://armature.tech/library/do-coding-agents-recommend-promptfoo) — Promptfoo was chosen in 13% of 288 judged evals sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend pytest-benchmark?](https://armature.tech/library/do-coding-agents-recommend-pytest-benchmark) — pytest-benchmark was chosen in 7% of 216 judged performance testing sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Railway?](https://armature.tech/library/do-coding-agents-recommend-railway) — Railway was chosen in 4% of 270 judged deployment sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Redis Query Engine?](https://armature.tech/library/do-coding-agents-recommend-redis-query-engine) — Redis Query Engine was chosen in 2% of 459 judged search sessions, ranking ninth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Redis?](https://armature.tech/library/do-coding-agents-recommend-redis) — Redis came up in 65% of judged cloud sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend Render?](https://armature.tech/library/do-coding-agents-recommend-render) — Render was chosen in 34% of 270 judged deployment sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Resend?](https://armature.tech/library/do-coding-agents-recommend-resend) — Resend is the most chosen email provider, taking 36% of 208 judged email sending sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Restate?](https://armature.tech/library/do-coding-agents-recommend-restate) — Coding agents raised Restate in 60 of 681 judged agent frameworks sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Retell AI?](https://armature.tech/library/do-coding-agents-recommend-retell-ai) — Retell AI was chosen in 18% of 308 judged voice agents sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Segment?](https://armature.tech/library/do-coding-agents-recommend-segment) — Segment was chosen in 5% of 359 judged product analytics sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend SendGrid?](https://armature.tech/library/do-coding-agents-recommend-sendgrid) — SendGrid was chosen in 11% of 208 judged email sending sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Sentry?](https://armature.tech/library/do-coding-agents-recommend-sentry) — Sentry is the most chosen observability tool, taking 37% of 360 judged observability sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Solr?](https://armature.tech/library/do-coding-agents-recommend-solr) — Coding agents raised Solr in 62 of 459 judged search sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Spring AI?](https://armature.tech/library/do-coding-agents-recommend-spring-ai) — Spring AI was chosen in 2% of 681 judged agent frameworks sessions, ranking number 12. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend SQLite?](https://armature.tech/library/do-coding-agents-recommend-sqlite) — SQLite came up in 90% of judged databases sessions and was adopted in none: on this board it is substrate, not a candidate. - [Do coding agents recommend Stripe?](https://armature.tech/library/do-coding-agents-recommend-stripe) — Stripe is the most chosen payment provider, taking 88% of 395 judged payments sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Supabase Auth?](https://armature.tech/library/do-coding-agents-recommend-supabase-auth) — Coding agents raised Supabase Auth in 69 of 201 judged authentication sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Supabase?](https://armature.tech/library/do-coding-agents-recommend-supabase) — Supabase was chosen in 3% of 356 judged databases sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Temporal?](https://armature.tech/library/do-coding-agents-recommend-temporal) — Temporal was chosen in 4% of 681 judged agent frameworks sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Turso?](https://armature.tech/library/do-coding-agents-recommend-turso) — Coding agents raised Turso in 105 of 356 judged databases sessions and chose it in none of them. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Twilio ConversationRelay?](https://armature.tech/library/do-coding-agents-recommend-twilio-conversationrelay) — Twilio ConversationRelay was chosen in 9% of 308 judged voice agents sessions, ranking sixth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Typesense?](https://armature.tech/library/do-coding-agents-recommend-typesense) — Typesense was chosen in 9% of 459 judged search sessions, ranking third. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Upstash?](https://armature.tech/library/do-coding-agents-recommend-upstash) — Upstash was chosen in 4% of 215 judged cloud sessions, ranking fifth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vapi?](https://armature.tech/library/do-coding-agents-recommend-vapi) — Vapi is the most chosen voice agent platform, taking 24% of 308 judged voice agents sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel AI Gateway?](https://armature.tech/library/do-coding-agents-recommend-vercel-ai-gateway) — Vercel AI Gateway was chosen in 16% of 140 judged AI gateways sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel AI SDK?](https://armature.tech/library/do-coding-agents-recommend-vercel-ai-sdk) — Vercel AI SDK is the most chosen agent framework, taking 14% of 681 judged agent frameworks sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel Analytics?](https://armature.tech/library/do-coding-agents-recommend-vercel-analytics) — Vercel Analytics was chosen in 8% of 359 judged product analytics sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel BotID?](https://armature.tech/library/do-coding-agents-recommend-vercel-botid) — Vercel BotID was chosen in 11% of 160 judged bot protection sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel Functions?](https://armature.tech/library/do-coding-agents-recommend-vercel-functions) — Vercel Functions was chosen in 23% of 287 judged serverless compute sessions, ranking second. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel Sandbox?](https://armature.tech/library/do-coding-agents-recommend-vercel-sandbox) — Vercel Sandbox was chosen in 14% of 299 judged code sandboxes sessions, ranking fourth. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend Vercel?](https://armature.tech/library/do-coding-agents-recommend-vercel) — Vercel is the most chosen hosting platform, taking 41% of 270 judged deployment sessions. Measured with Claude Code, Codex and Cursor. - [Do coding agents recommend WorkOS AuthKit?](https://armature.tech/library/do-coding-agents-recommend-workos-authkit) — WorkOS AuthKit is the most chosen authentication provider, taking 26% of 201 judged authentication sessions. Measured with Claude Code, Codex and Cursor. - [The same question, four different people, four different answers](https://armature.tech/library/how-personas-change-the-pick) — We asked for the same thing as a vibe coder, a junior, a senior engineer and an enterprise team. The category leader changed in 14 of 18 categories. - [The products coding agents consider and never choose](https://armature.tech/library/products-coding-agents-consider-and-never-choose) — 67 products reached the shortlist at least 25 times and were adopted zero times. Being rejected is cheaper to fix than being unknown. - [The dataset: which software coding agents choose](https://armature.tech/library/the-dataset) — 235 ranking rows across 18 categories from 5,292 judged sessions, as CSV and JSON, under CC BY 4.0. Cite it, load it, feed it to an agent. - [When coding agents build it themselves instead of buying](https://armature.tech/library/what-coding-agents-build-in-house) — In five of eighteen categories the most common outcome was hand-written code, not a product. Where agents build rather than adopt. - [What coding agents choose in a JavaScript project](https://armature.tech/library/what-coding-agents-choose-in-javascript) — 900 judged sessions inside JavaScript codebases. The leader differed from the overall leader in 2 of 7 categories. - [What coding agents choose in a Python project](https://armature.tech/library/what-coding-agents-choose-in-python) — 983 judged sessions inside Python codebases. The leader differed from the overall leader in 4 of 7 categories. - [What coding agents choose in a TypeScript project](https://armature.tech/library/what-coding-agents-choose-in-typescript) — 2,084 judged sessions inside TypeScript codebases. The leader differed from the overall leader in 3 of 15 categories. - [The repository decides more than your marketing does](https://armature.tech/library/what-decides-the-pick-the-repository) — The same request in a JavaScript and a TypeScript codebase produced different winners. An incumbent beat a written instruction. ## Playbooks One job, the steps to do it, and how to check that it worked. - [How to get picked for agent frameworks by coding agents](https://armature.tech/library/agent-frameworks-coding-agents-playbook) — Vercel AI SDK took 14% of 681 judged agent frameworks sessions. What the numbers say a vendor in this category should do. - [AGENTS.md and CLAUDE.md: what they actually do](https://armature.tech/library/agents-md-and-claude-md-what-they-actually-do) — A sentence in a repository instruction file was followed 16 times out of 18, and lost 2 of 3 once a rival library was installed. - [How to get picked for AI gateways by coding agents](https://armature.tech/library/ai-gateway-coding-agents-playbook) — Portkey took 21% of 140 judged AI gateways sessions. What the numbers say a vendor in this category should do. - [Audit your documentation for coding agents](https://armature.tech/library/audit-your-docs-for-coding-agents) — Twenty checks, each one testable in minutes, ordered by how much they change installs. Run this before you spend anything on content. - [How to get picked for authentication by coding agents](https://armature.tech/library/auth-coding-agents-playbook) — WorkOS AuthKit took 26% of 201 judged authentication sessions. What the numbers say a vendor in this category should do. - [How to get picked for bot protection by coding agents](https://armature.tech/library/bot-protection-coding-agents-playbook) — Cloudflare Turnstile took 57% of 160 judged bot protection sessions. What the numbers say a vendor in this category should do. - [How to choose a GEO or AI visibility agency](https://armature.tech/library/choosing-an-ai-visibility-agency) — Nine questions that separate an agency with a method from one with a dashboard subscription and a content calendar. Ask them before you sign anything. - [How to get picked for cloud by coding agents](https://armature.tech/library/cloud-coding-agents-playbook) — AWS took 62% of 215 judged cloud sessions. What the numbers say a vendor in this category should do. - [How to get picked for databases by coding agents](https://armature.tech/library/databases-coding-agents-playbook) — Neon took 66% of 356 judged databases sessions. What the numbers say a vendor in this category should do. - [How to get picked for deployment by coding agents](https://armature.tech/library/deploy-coding-agents-playbook) — Vercel took 41% of 270 judged deployment sessions. What the numbers say a vendor in this category should do. - [Developer relations metrics when half the audience is a machine](https://armature.tech/library/devrel-metrics-in-the-agent-era) — Talks, stars and impressions measured a human funnel. What to measure when adoption decisions no longer involve a person reading. - [How to get picked for evals by coding agents](https://armature.tech/library/evals-coding-agents-playbook) — Langfuse took 34% of 288 judged evals sessions. What the numbers say a vendor in this category should do. - [How to get your product recommended by Claude Code](https://armature.tech/library/how-to-get-recommended-by-claude-code) — Claude Code searches the web in 1.6% of decision runs and writes its own implementation in 19%. That changes which levers work. A specific playbook. - [How to get your product recommended by Codex](https://armature.tech/library/how-to-get-recommended-by-codex) — Codex searches the web in 53% of decision runs, which makes it the one agent where published content changes the outcome in weeks. - [How to get your product recommended by Cursor](https://armature.tech/library/how-to-get-recommended-by-cursor) — Cursor works with the project already open and indexed, which changes what it sees first. What moved its picks across 1,746 measured sessions. - [How to measure install share](https://armature.tech/library/how-to-measure-install-share) — The number that tells you whether coding agents install your product. How to measure it yourself in a weekend, and the traps that make a measurement worthless. - [How to show up in ChatGPT, Claude and Perplexity](https://armature.tech/library/how-to-show-up-in-chatgpt-claude-and-perplexity) — The three assistants retrieve differently, so the same page performs differently in each. What each one favours, and the work that pays across all three. - [How to win the enterprise persona](https://armature.tech/library/how-to-win-the-enterprise-persona) — Adding procurement language moved the winner to a Microsoft product in four unrelated categories. The mechanism is documentation. - [llms.txt: what it is, what it does, and what to put in it](https://armature.tech/library/llms-txt-guide) — About one site in ten has one. It takes an hour. Here is what the file is for, an honest account of what it does and does not do, and a template. - [How to get picked for email sending by coding agents](https://armature.tech/library/mail-coding-agents-playbook) — Resend took 36% of 208 judged email sending sessions. What the numbers say a vendor in this category should do. - [MCP server discoverability](https://armature.tech/library/mcp-server-discoverability) — How an agent finds, chooses and keeps using an MCP server. What to put in tool descriptions, why registries are not the main route yet, and what to measure. - [How to get picked for observability by coding agents](https://armature.tech/library/observability-coding-agents-playbook) — Sentry took 37% of 360 judged observability sessions. What the numbers say a vendor in this category should do. - [Open source as distribution, when the reader is an agent](https://armature.tech/library/open-source-as-distribution-for-agents) — A public repository was always readable, quotable and copyable. What changed is who copies from it, and how little they forgive. - [Package naming for coding agents](https://armature.tech/library/package-naming-for-coding-agents) — When the product name, the package name and the import name differ, an agent has to guess a mapping. Sometimes it installs somebody else's package. - [How to get picked for payments by coding agents](https://armature.tech/library/payments-coding-agents-playbook) — Stripe took 88% of 395 judged payments sessions. What the numbers say a vendor in this category should do. - [How to get picked for performance testing by coding agents](https://armature.tech/library/perf-ci-coding-agents-playbook) — JMH took 10% of 216 judged performance testing sessions. What the numbers say a vendor in this category should do. - [Writing a pricing page an agent can use](https://armature.tech/library/pricing-pages-agents-can-read) — Agents search for pricing more than you would expect, and they want units rather than plan names. Most pricing pages miss that. - [How to get picked for product analytics by coding agents](https://armature.tech/library/product-analytics-coding-agents-playbook) — PostHog took 53% of 359 judged product analytics sessions. What the numbers say a vendor in this category should do. - [Optimising a README and a repository for agents](https://armature.tech/library/readme-and-repository-optimization) — A public repository is a first-class artefact for an agent. What to put in it, and the parts that decide whether your code is copied. - [robots.txt for AI crawlers](https://armature.tech/library/robots-txt-for-ai-crawlers) — Which AI user agents exist, what each one is for, and why blocking the wrong one removes you from the answers instead of protecting you. - [How to get picked for code sandboxes by coding agents](https://armature.tech/library/sandboxes-coding-agents-playbook) — E2B took 42% of 299 judged code sandboxes sessions. What the numbers say a vendor in this category should do. - [How to get picked for search by coding agents](https://armature.tech/library/search-coding-agents-playbook) — Postgres Full-Text Search took 17% of 459 judged search sessions. What the numbers say a vendor in this category should do. - [How to get picked for serverless compute by coding agents](https://armature.tech/library/serverless-coding-agents-playbook) — AWS Lambda took 24% of 287 judged serverless compute sessions. What the numbers say a vendor in this category should do. - [How to get picked for file storage by coding agents](https://armature.tech/library/storage-coding-agents-playbook) — Amazon S3 took 46% of 90 judged file storage sessions. What the numbers say a vendor in this category should do. - [Time to first successful call, when the developer is an agent](https://armature.tech/library/time-to-first-successful-call) — The oldest onboarding metric, with a new reader. An agent does not persist through friction, it switches. What that changes about where you measure. - [How to get picked for voice agents by coding agents](https://armature.tech/library/voice-agents-coding-agents-playbook) — Vapi took 24% of 308 judged voice agents sessions. What the numbers say a vendor in this category should do. - [How to write a quickstart an agent can follow](https://armature.tech/library/write-a-quickstart-an-agent-can-follow) — The highest return change most developer tool companies can make. What breaks, why it costs installs, and the exact shape that works. - [How to write a page an AI answer will quote](https://armature.tech/library/write-content-an-ai-answer-will-quote) — Extraction rewards one specific shape: a question heading, then a self-contained answer in 40 to 60 words. Most pages bury the answer in paragraph four. ## Answers Short answers to the questions people actually ask. - [Can you influence what AI recommends?](https://armature.tech/library/can-you-influence-what-ai-recommends) — Yes, and the levers are measurable and ranked. Here is what actually moves the outcome, in order of size, and the two things that do not work. - [Do AI coding assistants actually choose which tools to install?](https://armature.tech/library/do-ai-coding-assistants-choose-which-tools-to-install) — Yes, routinely, and often without a person comparing anything. What the data shows about how often it happens and how concentrated the outcomes are. - [Does the programming language change which tools AI picks?](https://armature.tech/library/does-the-programming-language-change-what-ai-picks) — Yes, more than anything else we measured. The same request produced one winner in 111 of 111 JavaScript sessions and a four-way split on TypeScript. - [Is GEO worth it for a developer tool company?](https://armature.tech/library/is-geo-worth-it-for-a-developer-tool) — Partly, and the honest answer depends on one number you can measure: how many of your users run an agent that searches the web at all. - [Which coding agent should you optimise for?](https://armature.tech/library/which-coding-agent-should-i-optimise-for) — They need different work. Codex reads the web in half its decisions, Claude Code in 1.6%. The answer depends on one thing you can find out about your own users. - [Why did the coding agent pick my competitor?](https://armature.tech/library/why-did-the-agent-pick-my-competitor) — Five reasons, ranked by how often they occur in judged sessions. Four of the five are defects you can fix in days, not preferences you have to out-argue. ## Glossary One page per term, so the words mean the same thing to everyone. - [Agent readiness](https://armature.tech/library/agent-readiness) — Whether a product can be found, understood and correctly integrated by a coding agent without a person filling the gaps. A checklist, not a score. - [AI Overviews](https://armature.tech/library/ai-overviews) — Google's generated answer above the results. What gets pulled into one, what does not, and why a developer tool should not build a strategy on it alone. - [AI visibility](https://armature.tech/library/ai-visibility) — How often and how favourably AI systems mention a brand when they answer questions. What it measures, how it is tracked, and what it misses. - [Coding agent](https://armature.tech/library/coding-agent) — A tool that reads a codebase, decides what to do, and writes code. Which ones matter, and why the differences change what gets installed. - [Consideration rate](https://armature.tech/library/consideration-rate) — How often an agent raised your product as a candidate versus how often it chose it. Mailgun was raised in 97 email sessions and chosen in none. - [Generative engine optimization (GEO)](https://armature.tech/library/generative-engine-optimization) — The practice of making content that AI systems use and cite when they write answers. What the term means, how it is measured, and where it stops. - [Harness](https://armature.tech/library/harness) — The program around the model in a coding agent: how it reads files, which tools it has, how it plans. It decides which lever works. - [In-house rate](https://armature.tech/library/in-house-rate) — The share of sessions where a coding agent wrote the implementation itself instead of adopting any product. It ranged from 0% to 52% by category. - [Install share](https://armature.tech/library/install-share) — The percentage of sessions in a category where a coding agent picked your product and wrote working integration code. The agent-era equivalent of market share. - [Judge model](https://armature.tech/library/judge-model) — The model that reads a session and decides what was chosen. Why it must come from a different family, and the rules that matter. - [LLMO (large language model optimization)](https://armature.tech/library/llmo) — A third name for the same work as GEO and AEO. What the term adds, what it does not, and the one distinction that is actually load bearing. - [Markdown twin](https://armature.tech/library/markdown-twin) — The same page served as Markdown at the same address plus .md. Agents read it far better than a rendered site, and almost nobody publishes one. - [Model Context Protocol (MCP)](https://armature.tech/library/model-context-protocol) — The open standard for connecting AI clients to tools and data. What it is, why it changed developer tool distribution, and what it means for discoverability. - [Open seam](https://armature.tech/library/open-seam) — The thing a test repository is missing, which is what makes it a valid test. A repository that already has your category cannot measure choice. - [Persona prompt](https://armature.tech/library/persona-prompt) — The same request written as four kinds of person. It changed the category leader in 14 of 18 categories, which is the cheapest split. - [Programmatic SEO, and when it becomes doorway pages](https://armature.tech/library/programmatic-seo) — Generating many pages from one data set works when each page carries data only you have. It fails when they are one page renamed. - [Prompt monitoring](https://armature.tech/library/prompt-monitoring) — Sending a fixed list of prompts to AI assistants on a schedule and counting brand mentions. How every AI visibility tool works, and what the method cannot see. - [Self-preference](https://armature.tech/library/self-preference) — Whether a coding agent favours its own maker's products. One category showed it, seven did not. What we found and how we controlled for it. - [Share of voice (in AI answers)](https://armature.tech/library/share-of-voice) — Your brand mentions as a proportion of all brand mentions across a tracked prompt set. Useful against your own history, meaningless as an absolute number. - [Structured data for a developer tool site](https://armature.tech/library/structured-data-for-developer-tools) — Which schema types are worth adding, which are a waste, and the one rule that turns markup from an asset into a manual penalty.