bidirectional-bridge-claude-codex: Skill for Claude Code

.claude/skills/using-bridge/SKILL.md

using-bridge is a skill for Claude Code, Codex from grizzly2005/bidirectional-bridge-claude-codex. It costs 81 tokens per session (2,309 once invoked), scanned A, original, MIT.

A bridge for coordinating Claude Code and Codex through a repository-local MCP service. It supports bounded delegation, independent review, recovery of interrupted work, and event observation.

In plain words
What is it for?
Use it to delegate an independent subtask, request a second review, resume a stranded task, or inspect bridge activity when that coordination is needed.
Why use it?
It allows another coding-agent runtime to handle a clearly defined part of substantial implementation, diagnosis, review, or recovery work.

Skill for Claude CodeCodex

Written for Claude Code and Codex: installed under .claude/, but also agents/openai.yaml present. Also seen: mentions Claude Code; mentions Codex.

This is grizzly2005/bidirectional-bridge-claude-codex's own configuration. It tells Claude Code and Codex how to work on bidirectional-bridge-claude-codex itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything bidirectional-bridge-claude-codex configures →

Reuse

Borrowing it

Nothing to install: this file belongs to grizzly2005/bidirectional-bridge-claude-codex. 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/grizzly2005/bidirectional-bridge-claude-codex/main/.claude/skills/using-bridge/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/grizzly2005/bidirectional-bridge-claude-codex

Made for: Claude Code, Codex.

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.

agentmods badge for using-bridge

README.md
[![agentmods](https://agentmods.dev/badge/skills/grizzly2005/bidirectional-bridge-claude-codex/using-bridge.svg)](https://agentmods.dev/skills/grizzly2005/bidirectional-bridge-claude-codex/using-bridge)
Your own site
<a href="https://agentmods.dev/skills/grizzly2005/bidirectional-bridge-claude-codex/using-bridge"><img src="https://agentmods.dev/badge/skills/grizzly2005/bidirectional-bridge-claude-codex/using-bridge.svg" alt="Measured on agentmods" height="20"></a>
Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,309 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00081 $0.02309
Opus 5 $0.00041 $0.01154
Sonnet 5 $0.00016 $0.00462
Haiku 4.5 $0.00008 $0.00231

Measured 7d ago against content hash 52e51c8b4c08, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

using-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 7d 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.

.claude/skills/using-bridge/SKILL.md · 186 lines

How it starts

The opening of the file, as written. The whole thing — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Use the Bridge

Use the bridge as a coordination layer, not as a reason to involve another agent. Keep the current native client responsible for the user request; delegate only a bounded subtask whose expected value exceeds startup, context, and verification overhead.

Choose the path

  • Delegate: assign one independent, verifiable subtask to the other runtime.
  • Review: ask the other runtime for a bounded, usually read-only, independent review.
  • Recover: resume one known stranded task in place after quota, timeout, crash, or process interruption.
  • Observe: inspect telemetry or events only when the user requests them, during dogfooding, or for diagnosis.
  • Do not use the bridge: complete trivial, tightly coupled, or cheaper single-agent work locally.

If this session is already a delegated worker with a supplied task_id, scope, expected deliverable, and verification criteria, complete that contract. Do not create a new root or delegate again unless the contract explicitly permits it.

Decide whether to delegate

Delegate only when all are true:

  1. The child task is independently understandable from a compact contract.
  2. Its write scope is disjoint, or it is read-only.
  3. Its output and verification criteria are explicit.
  4. The other runtime adds distinct implementation, diagnosis, review, or research value.
  5. The manager can consume and verify the result without prolonged back-and-forth.

Do not delegate:

  • a trivial command, fact lookup, or tiny edit the manager can finish immediately;
  • work requiring continuous conversation between agents;
  • overlapping writes to the same files or module;
  • ambiguous work missing authority, scope, or acceptance criteria;
  • work with no meaningful verification;
  • merely to collect telemetry or “use both models”;
  • when the user forbids the bridge.

Treat an exact user-requested delegation count as a hard limit.

Delegation checkpoint for substantial work

For every substantial request, determine whether one bounded child task would improve implementation quality, security, diagnosis, verification, or independent review.

Read the full file on GitHub · 186 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 7d ago First seen · 186 lines · 81 tokens per session scan A 52e51c8b4c08

Subscribe to this mod's changes

using-bridge is a skill published in the GitHub repository grizzly2005/bidirectional-bridge-claude-codex (1 stars, last pushed 25d ago), licensed MIT. It adds 81 tokens to every session and 2,309 once invoked, about $0.0004 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.