What is claude-science-api-bridge AGENTS.md?

A set of instructions for an AI coding agent working on a local bridge between Claude Science and OpenAI-compatible services.

It is for configuring supported API providers, local proxy components, compatible clients, and image-capable models while protecting network settings and secrets.

It gives the agent repository guidance and safety limits for changing network and proxy settings.

How Jyx0208/claude-science-api-bridge configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to Jyx0208/claude-science-api-bridge. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/Jyx0208/claude-science-api-bridge/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/Jyx0208/claude-science-api-bridge

Made for: Codex, OpenCode.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/jyx0208/claude-science-api-bridge/agents-md/github.svg)](https://agentmods.dev/instructions/jyx0208/claude-science-api-bridge/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/jyx0208/claude-science-api-bridge/agents-md"><img src="https://agentmods.dev/badge/instructions/jyx0208/claude-science-api-bridge/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/jyx0208/claude-science-api-bridge/agents-md"><img src="https://agentmods.dev/badge/instructions/jyx0208/claude-science-api-bridge/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,090 This file is loaded in full into every session.

When invoked 1,090 The same file — it is already loaded in full.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.01090	$0.01090
Opus 5.5	$0.00436	$0.00436
Sonnet 5.5	$0.00218	$0.00218
Haiku 4.5	$0.00109	$0.00109

Measured 2mo ago against content hash 54fa7110a0b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

claude-science-api-bridge AGENTS.md scanned grade A with 1 finding 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 2mo 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -sS http://127.0.0.1:9876/health

AGENTS.md · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Operating Manual

This repository is designed so an AI coding agent can configure Claude Science to use an OpenAI-compatible third-party API through a local Anthropic-compatible proxy.

Read this file first, then follow docs/agent-runbook.md.

Prime Directive

Do not break the user's network.

Default to safe mode:

Do not edit Clash, Surge, Shadowrocket, system proxy, VPN, DNS, or TUN settings.
Do not reload network daemons.
Do not write to /etc/hosts.
Do not install a root CA.
Do not bind port 443.
Do not print, commit, or summarize API keys, OAuth tokens, private keys, or certificate private keys.

Only use advanced HTTPS interception after the user explicitly approves it for the current machine.

Goal

Make Claude Science usable with DeepSeek, OpenAI, or another OpenAI-compatible API provider. If the user needs image understanding, choose a vision-capable backend model and preserve image inputs instead of replacing them with text placeholders.

On Linux, support covers the local proxy, Dashboard, service installation, and compatible clients that honor ANTHROPIC_BASE_URL. Claude Science desktop startup and daemon patches are macOS-only.

The safe path is:

Run a local HTTP proxy on 127.0.0.1:9876.
Set ANTHROPIC_BASE_URL=http://127.0.0.1:9876.
Generate a local fake Claude Science OAuth token.
Configure an API key and model mapping in config.json or the dashboard.
Configure model_aliases and model_list_mode=aliases so Claude Science can show third-party model names.
Choose *_upstream_mode=anthropic for providers with native Anthropic endpoints; otherwise use openai.
Set inline_image_policy=preserve or auto only when the selected backend supports image input.
Optionally enable proxy_auth_mode=required only when the launch path will include the secret.
Start or restart Claude Science with scripts/start-claude-science.sh.
Verify /v1/models and /v1/messages reach the proxy and the backend succeeds.

Read the full file on GitHub · 93 lines

Changes

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.

2mo ago First seen · 93 lines · 1,090 tokens per session scan A 54fa7110a0b0

Subscribe to this mod's changes

claude-science-api-bridge AGENTS.md is an instructions file published in the GitHub repository Jyx0208/claude-science-api-bridge (52 stars, last pushed 4mo ago), licensed MIT. It adds 1,090 tokens to every session, about $0.0044 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

anthropic claude-science deepseek macos openai-compatible proxy

claude-science-api-bridge AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
claude-science-api-bridge AGENTS.mdJyx0208/claude-science-api-bridge	Instructions file	52	4mo ago	MIT	A	1,090	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	yesterday	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

claude-science-api-bridge: Instructions file for Codex