agent-bridge

A bridge that lets Codex delegate code changes in a Git repository to an external coding agent such as Claude Code or DeepSeek Harness, then review the result.

In plain words
What is it for?
Use it to run an external agent on repository tasks, respond to its questions, inspect verification results, continue or reject work, and apply approved changes.
Why use it?
It separates supervision from implementation and provides a process for checking, accepting, rejecting, or applying the external agent's changes.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/jaylanjerry/codex-agent-bridge/agent-bridge
Any agent
npx skills add JaylanJerry/codex-agent-bridge --skill agent-bridge
Clone the repo
git clone --depth 1 https://github.com/JaylanJerry/codex-agent-bridge

Made for: Claude Code, Codex.

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,144 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00057 $0.01144
Opus 5 $0.00028 $0.00572
Sonnet 5 $0.00011 $0.00229
Haiku 4.5 $0.00006 $0.00114

Measured 2d ago against content hash 532b063827b7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-bridge 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 2d 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.

skills/agent-bridge/SKILL.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Bridge

Codex 是 Supervisor。Worker 是外部完整 Coding Agent。Bridge 不调用 LLM

MCP server 名:agent-bridge
工具前缀:bridge_*

安装:npx -y codex-agent-bridge,然后新开 Codex。

什么时候用

用户要改一个 git 仓库,且应该由 Claude Code 或 DeepSeek Harness 执行。

什么时候不用

  • 只问问题、读代码、做设计
  • 没有 git 仓库
  • OpenCode(当前 DEFERRED)
  • Codex 自己改当前工作区即可

闭环

bridge_run
→ 若 state=WAITING_FOR_INPUT:读 pendingInput.options,bridge_respond(optionId + stateVersion)
→ 若 state=TASK_TIMED_OUT:wait 预算到了,Worker 已停,worktree 保留。读 ReviewPacket,然后 continue 或 reject
→ 重复直到 AWAITING_REVIEW
→ 读 structuredContent.reviewPacket(含 verification)
→ verification.passed=false 时不要 approve,除非用户明确要求
→ 不通过:bridge_continue(notes + stateVersion)
→ 再读 ReviewPacket(continue 过程中也可能再 WAITING_FOR_INPUT)
→ 通过:bridge_approve(stateVersion)
→ 用户要落到当前分支时:bridge_apply(cherry-pick,不是 merge)

worker:调试用 replay;闸门自检可用 fake;真干活用 claudedeepseek。V1 只选 worker。不要传、不要编造 model / provider / effort;用户已在 Claude Code 或 DeepSeek Harness 里按原生方式配好,Bridge 启动时继承。Claude 背后的官方 Anthropic、CC Switch 或其他第三方 Provider 是 Worker 内部实现,不要猜测 effective upstream model。

项目里如果有 .agent-bridge/verify.jsonbridge_run 会默认跑其中全部 verifyId。也可显式传 verifyIds。示例:templates/verify.json

破坏性调用必须带当前 task.stateVersion。冲突时不要重试旧版本。

每次 bridge_* 任务调用都必须带同一个 project 路径。任务 UUID 的字段名是 task,不是 taskIdbridge_respond 示例:project + task + stateVersion + optionId。漏掉 project 会查到空任务库或报 journal 不存在,看起来像状态丢失。

clientRequestId 相同且内容相同会返回同一 task;内容不同会 TASK_ALREADY_EXISTS

MCP 默认 permissionMode=gate:Worker 要写文件/跑命令时会停在 WAITING_FOR_INPUT,不要自动当任务完成。选 allow_once,除非用户明确禁止。optionId 以 pendingInput.options 为准(Claude 可能是 allow 而不是 allow-once)。没有 live waiter 时(Core 重启)不要 respond,改 bridge_continue

timeoutMs 到期会停 Worker、保留 worktree,状态变成 TASK_TIMED_OUT。这不是完成。可以 bridge_continuebridge_reject

CLI 默认 auto(一次进程无法跨调用停闸)。

掉线后先 bridge_statusneedsAttention: true)或 bridge_doctor。Core 重启会把 QUEUED / STARTING / RUNNING / VERIFYING / WAITING_FOR_INPUT 收成 AWAITING_REVIEW + interrupted,不要 respond,改 bridge_continue(Claude 可尝试 session/load)。doctor 标出遗留 worktree 时用 bridge_prune

Read the full file on GitHub · 71 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.

  1. 2d ago First seen · 71 lines · 57 tokens per session scan A 532b063827b7

Subscribe to this mod's changes

agent-bridge is a skill published in the GitHub repository JaylanJerry/codex-agent-bridge (0 stars, last pushed 10d ago), licensed MIT. It adds 57 tokens to every session and 1,144 once invoked, about $0.0003 per session on Opus 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 2026-08-31.

Related

Other skills, from other repositories

vision-skills

Local vision CLIs: glance (describe/ask/OCR an image), ground (locate a target, pixel box), detect (element inventory), trace (image to SVG geometry), crop (cut a pixel box to a file), and scripts/htmlshot.py (HTML file to image). Use for any task involving an image — questions, text, splitting and transcribing long…

Anionex/agent-vision-toolkit · 132 tokens

dsh-doc-site-sync

Use when publishing, updating, moving, or removing DeepSeek Harness documentation website pages; editing website/docs.ts mappings or navigation; diagnosing a page missing from the VitePress site; fixing projected documentation links; or running the docs:dev, docs:check, and doc-sync workflow after website-content…

whitelonng/dshcode · 67 tokens

creating-workflows

当用户想「创建/生成一个 workflow(可复用的多子-agent 编排脚本)」时使用。讲清 deepx workflow 的 JavaScript 脚本格式、可用 API,以及如何用 Workflow 工具保存与运行。.

itmisx/deepx-code · 58 tokens

subagent-driven-development

Execute plan by dispatching fresh subagent per task, with two-stage review after each: spec compliance review first, then code quality review.

itmisx/deepx-code · 37 tokens

standard

Apply the standard Coding workspace workflow with explicit inspection, scoped edits, and proportionate validation.

zhnt/loushang · 20 tokens

openspec

规格驱动开发(SDD)。实现任何功能或改动时使用——写代码前先就规格达成一致:在 openspec/changes/ / 写 proposal、spec deltas、tasks(复杂时加 design),对齐后再严格按规格实现,完成后把 delta 归档进基线 specs/。.

itmisx/deepx-code · 73 tokens