NeuroAPI is an AI API service that provides access to OpenAI, Anthropic Claude, Google Gemini, DeepSeek, image-generation, and video-generation models through one account. It is for users of Codex CLI and Claude Code who want to connect those tools through NeuroAPI on Windows or macOS. The catalogue instruction helps install and configure those clients for the service.
Borrowing it
Nothing to install: this file belongs to neurogen-dev/NeuroAPI. 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/neurogen-dev/NeuroAPI/agents/AGENTS.mdgit clone --depth 1 https://github.com/neurogen-dev/NeuroAPIWrote 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/neurogen-dev/neuroapi/agents-md)<a href="https://agentmods.dev/instructions/neurogen-dev/neuroapi/agents-md"><img src="https://agentmods.dev/badge/instructions/neurogen-dev/neuroapi/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/neurogen-dev/neuroapi/agents-md"><img src="https://agentmods.dev/badge/instructions/neurogen-dev/neuroapi/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 | $0.00509 | $0.00509 |
| Opus 5.5 | $0.00204 | $0.00204 |
| Sonnet 5.5 | $0.00102 | $0.00102 |
| Haiku 4.5 | $0.00051 | $0.00051 |
Grade A, and why
NeuroAPI 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 23d 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.
What it actually says
NeuroAPI Agents repository guidance
-
This branch is the public, auditable installer for NeuroAPI integrations with Codex CLI and Claude Code.
-
Never accept API keys through command-line arguments, persist them as plaintext, or print them outside the dedicated credential helpers.
-
Windows secrets must stay DPAPI-protected for the current user. macOS secrets must stay in the current user's login Keychain.
-
Do not overwrite unrelated Codex, Claude Code, shell, or PATH configuration. Update and uninstall only files marked as installer-owned.
-
Keep Windows PowerShell 5.1 compatibility and macOS system Bash compatibility.
-
Validate script syntax, generated TOML/JSON, idempotency, secret handling, and uninstall boundaries before publishing.
-
Codex CLI uses a user-level profile-v2 file,
https://codex.neuroapi.host/v1, and WebSocket by default; the optional Codex Desktop setup uses the same host with HTTP/SSE. Publish installers only after authenticated catalog and transport checks. -
Managed Codex CLI uses verified hosted
web_search = "live",plugins = falseandremote_plugin = false: API-key startup must not sync plugin catalogs. Standalone user-configured MCP remains independent. Multi-agent namespace, goals, apps and browser use stay disabled in the CLI profile until independently verified. Do not advertise standalone web-search support. Desktop edits onlyremote_plugin, preserving existingpluginsand other feature preferences; document optional manual plugin disabling. -
The validated Codex catalog
default_modelis authoritative for setup and launch on both platforms, including Desktop; never hardcode a preferred model or infer the default from list order. -
Managed launchers fetch and validate fresh key-scoped catalogs into private per-launch snapshots; never fall back to stale/bundled lists or accept executable server settings. Claude Code's base URL is
https://claude.neuroapi.host; Claude Desktop gateway is configured separately in its UI. Keep ordinary keys, preserve unrelated configuration, and document managed-policy/explicit-override boundaries. -
Claude client-settings v2 is opt-in via
X-NeuroAPI-Client-Settings-Version: 2; accept only the three reviewed limit/hint env keys with canonical bounded decimal strings (or hint1), retain the 4096 output fallback for old servers.--doctorperforms no generation;--doctor-generateexplicitly opts into one bounded HTTP probe with no retries, secret-free output and no claim of WebSocket/tool compatibility.
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.
- 23d ago Changed · +1 lines · +89 tokens per session f59be7973656
- 25d ago Changed · +2 lines · +94 tokens per session 486f7aa01f09
- 27d ago Changed · +22 tokens per session e1beca2587cb
- 28d ago Changed · +1 lines · +45 tokens per session f93732baa78a
- 1mo ago Changed · +3 lines · +111 tokens per session aaf6cde9228c
- 1mo ago First seen · 9 lines · 148 tokens per session scan A 83ddb90856d5
NeuroAPI AGENTS.md is an instructions file published in the GitHub repository neurogen-dev/NeuroAPI (1,317 stars, last pushed 24d ago), licensed MIT. It adds 509 tokens to every session, about $0.0020 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.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.