Skip to content

giladbarnea/soft-skills / plugins/soft-skills/skills/ai-to-delegated

ai-to-delegated

Verified

Best practices for getting a step-function leap in productivity when working with other AI’s. Load `ai-to-delegated` before spawning a sub-agent, creating a team, talking to a teammate agent, or writing a doc likely read by other agents. Trigger words include “dispatch an agent”, “spawn a team”, and “fork the session to ...”.

planning/dispatch-outside-the-harness.md

View on GitHub
---
name: dispatch-outside-the-harness
description: Start a delegate with a terminal multiplexer when the harness cannot set the model, thinking level, and context combination you chose.
---

# Dispatch Outside the Harness

Read this when the configuration you chose in [choose-configuration.md](planning/choose-configuration.md) is one your harness cannot produce, and the user is willing to use a terminal multiplexer.

Proprietary harnesses limit exactly this. The main agent cannot pick any {model × thinking × context fork} combination it wants. Claude Code and Codex both have this limit.

The bypass is a terminal multiplexer the user already runs, such as `herdr`, `tmux`, or `cmux`. Open a pane per delegate and start the session yourself with the CLI flags below. Then message that session through your harness's session-to-session messaging if it has one. If it does not, communicate through the multiplexer.

## CLI semantics

**Claude Code**

```text
claude --model MODEL --effort THINKING_LEVEL
```

To fork an existing session, add:

```text
--fork-session --resume SESSION_ID
```

**Codex**

```text
codex --model MODEL --config model_reasoning_effort=THINKING_LEVEL
```

The `=` in `--config` is required. Forking is a subcommand:

```text
codex fork SESSION_ID [--model MODEL] [--config model_reasoning_effort=THINKING_LEVEL]
```
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall