Ax is a TypeScript-first programming framework for building applications with large language models through typed generation, agents, workflows, and optimization tools. It is intended for developers who want one model for LLM programs across TypeScript, Python, Java, C++, Go, Rust, and other runtimes.
Borrowing it
Nothing to install: this file belongs to ax-llm/ax. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ax-llm/ax/main/AGENTS.mdgit clone --depth 1 https://github.com/ax-llm/axWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/ax-llm/ax/agents-md)<a href="https://agentmods.dev/instructions/ax-llm/ax/agents-md"><img src="https://agentmods.dev/badge/instructions/ax-llm/ax/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/ax-llm/ax/agents-md"><img src="https://agentmods.dev/badge/instructions/ax-llm/ax/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01582 | $0.01582 |
| Opus 5.5 | $0.00633 | $0.00633 |
| Sonnet 5.5 | $0.00316 | $0.00316 |
| Haiku 4.5 | $0.00158 | $0.00158 |
Grade A, and why
ax AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ax Repository Guide
What This File Is For
This file should stay short.
Use it for repo-specific rules, workspace conventions, and pointers to the deeper subsystem docs in src/ax/skills/.
Do not turn this file into a second full API manual.
Repository Layout
src/ax/- main library package:@ax-llm/axsrc/ai-sdk-provider/- Vercel AI SDK provider:@ax-llm/ax-ai-sdk-providersrc/examples/- runnable examples and integration-style demoswebsite/- Hugo documentation site, deployed to axllm.dev via GitHub Pagesdocs/- canonical markdown docs
Canonical Docs
docs/ holds the canonical maintainer markdown. The public site under
website/ is built separately with npm run website:build (TypeDoc API
markdown → npm run website:prepare → Hugo → Pagefind); never edit the
generated copies under website/.generated/ or website/public/.
Current Project Defaults
Prefer the modern factory-style API:
- use
ai(...),ax(...),agent(...),flow(...),optimize(...) - prefer string signatures and
s(...) - prefer
fn(...)for tools
Avoid deprecated patterns in new code:
new AxAI(...)- template literal forms like
ax`...`ands`...`
Use current models
When adding or updating any example, default, or test that names a model, pick a current
generation — newer beats biggest. The provider model enums are the source of truth for what's
current: src/ax/ai/openai/chat_types.ts, src/ax/ai/google-gemini/types.ts,
src/ax/ai/anthropic/types.ts. Choose a recent member; a newer small/cheap model
(e.g. gpt-5.4-mini, gemini-3.5-flash) is preferred over an old flagship (gpt-4o,
gemini-2.x), and never pin a dated -preview when a stable newer one exists. Don't leave
examples sitting on an old generation once a newer one lands in the enums.
Skills As Subsystem Docs
Use the skill files in src/ax/skills/ as the primary subsystem documentation:
src/ax/skills/ax-llm.md- top-level Ax quick referencesrc/ax/skills/ax-ai.md- AI providers, models, presets, embeddings, thinkingsrc/ax/skills/ax-typesafe.md- Typesafe/Jev signatures, native questions, criteria, scoring, and hybrid generation (TypeScript)src/ax/skills/ax-audio.md- conversational audio input/output in.chat()src/ax/skills/ax-signature.md- signatures, field types, validation, fluent APIsrc/ax/skills/ax-gen.md- generators and structured outputsrc/ax/skills/ax-agent.md- agents, runtime, discovery, delegation, shared fieldssrc/ax/skills/ax-agent-optimize.md-agent.optimize(...), evals, judges, artifactssrc/ax/skills/ax-flow.md- workflows and orchestrationsrc/ax/skills/ax-mcp.md- native MCP clients, transports, auth, catalogs, tasks, subscriptions, and Ax integrationsrc/ax/skills/ax-event-runtime.md- event ingestion, wake/resume, continuations, stores, and sinkssrc/ax/skills/ax-gepa.md- Pareto optimizationsrc/ax/skills/ax-learn.md- self-improving agent patterns
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · +1 lines · +34 tokens per session 50671c93cb6c
- 2mo ago First seen · 150 lines · 1,548 tokens per session scan A e216ef241ab6
ax AGENTS.md is an instructions file published in the GitHub repository ax-llm/ax (2,978 stars, last pushed 3d ago), licensed Apache-2.0. It adds 1,582 tokens to every session, about $0.0063 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
ax AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| ax AGENTS.mdax-llm/ax | Instructions file | 3.0k | 3d ago | Apache-2.0 | A | 1,582 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |