What is delegate-to-peer?

A skill for assigning a bounded, independent task from one AI coding agent to another through Agent Bridge.

Delegating work between Claude, Codex, and ChatGPT, retrieving the result, and checking it before use.

A second agent can provide a separate implementation, review, research result, or validation while the first agent continues its work.

How jcharveyjr/claude-chatgpt-agent-bridge configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to jcharveyjr/claude-chatgpt-agent-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/jcharveyjr/claude-chatgpt-agent-bridge/main/.claude/skills/delegate-to-peer/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/jcharveyjr/claude-chatgpt-agent-bridge

Made for: Claude Code.

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/skills/jcharveyjr/claude-chatgpt-agent-bridge/delegate-to-peer/github.svg)](https://agentmods.dev/skills/jcharveyjr/claude-chatgpt-agent-bridge/delegate-to-peer)

Your own site

<a href="https://agentmods.dev/skills/jcharveyjr/claude-chatgpt-agent-bridge/delegate-to-peer"><img src="https://agentmods.dev/badge/skills/jcharveyjr/claude-chatgpt-agent-bridge/delegate-to-peer/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/skills/jcharveyjr/claude-chatgpt-agent-bridge/delegate-to-peer"><img src="https://agentmods.dev/badge/skills/jcharveyjr/claude-chatgpt-agent-bridge/delegate-to-peer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 78 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 549 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.

Model	Per session	Once invoked
Fable 5.1	$0.00078	$0.00549
Opus 5.5	$0.00031	$0.00220
Sonnet 5.5	$0.00016	$0.00110
Haiku 4.5	$0.00008	$0.00055

Measured 2mo ago against content hash 0534a49a22b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

delegate-to-peer 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 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.

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/delegate-to-peer/SKILL.md · 41 lines

How it starts

The opening of the file, as written. The whole thing — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Delegate to Peer

Use the Agent Bridge MCP tools by their base names even when the client prefixes them.

Call agent_bridge_capabilities before the first delegation in a session.
Choose the peer target:
- From Claude, set source_agent to claude and target_agent to codex.
- From Codex, set source_agent to codex and target_agent to claude.
- From ChatGPT Work, set source_agent to chatgpt and choose claude for general peer work or codex for coding-focused work.
Delegate one bounded outcome with delegate_task. Include the workspace name, necessary context, explicit success criteria, and the least-permissive mode.
Continue independent work while the task is queued or running. Do not busy-wait.
Call get_task when the peer result is needed. If the task is not finished, wait for a meaningful work boundary before checking again.
Verify the returned work before integrating it. The requesting agent remains responsible for the final answer and validation.

Guardrails

Never delegate to the same agent as source_agent.
Never ask the worker to call Agent Bridge again. The broker also injects this restriction.
Pass parent_task_id for a true nested task. Do not evade the configured depth limit.
Use read_only for analysis, review, planning, or research. Use workspace_write only when file edits are required.
Do not send passwords, tokens, private keys, session cookies, or unrelated private data in task text or context.
Do not delegate purchases, messages, publishing, deployments, permission changes, or destructive actions without the user's explicit authority and the destination runtime's own approval.
Cancel obsolete work with cancel_task.

Good Delegation Shape

Provide:

One concrete deliverable
Only the context needed for that deliverable
Named workspace rather than an arbitrary filesystem path
Observable success criteria
Expected handoff, including files changed and validation performed

Read the full file on GitHub · 41 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 · 41 lines · 78 tokens per session scan A 0534a49a22b9

Subscribe to this mod's changes

delegate-to-peer is a skill published in the GitHub repository jcharveyjr/claude-chatgpt-agent-bridge (1 stars, last pushed 3mo ago), licensed MIT. It adds 78 tokens to every session and 549 once invoked, about $0.0003 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 2026-09-01.

Building agents

delegate-to-peer next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
delegate-to-peerjcharveyjr/claude-chatgpt-agent-bridge	Skill	1	3mo ago	MIT	A	78	Claude Code
imagegenopenai/codex	Skill	134k	today	Apache-2.0	A	113	Codex
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	yesterday	MIT	A	200	Claude Code, Codex
playwright-climicrosoft/playwright	Skill	100k	yesterday	Apache-2.0	A	19	Claude Code
browser-automationopenclaw/openclaw	Skill	394k	today	MIT	A	36	Claude Code
systematic-debuggingobra/superpowers	Skill	314k	4d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	95	Claude Code

claude-chatgpt-agent-bridge: Skill for Claude Code