herdr-toolkit
llms.txt0 starsChanged 33 days agoMITView on GitHub
Claude Code plugin with two operating skills for running Claude Code on top of Herdr, the terminal agent multiplexer: spawning detached Remote Control sessions into Herdr panes (spawn-session), and sending prompts to agents in Herdr panes with landing confirmation and a completion wait (agent-send).
In this repository3 other files it gives its agents.
- llms-full.txtllms.txt
- agent-sendSkill
- spawn-sessionSkill
The file
Files
# herdr-toolkit > Claude Code plugin with two operating skills for running Claude Code on top of Herdr, the terminal agent multiplexer: spawning detached Remote Control sessions into Herdr panes (spawn-session), and sending prompts to agents in Herdr panes with landing confirmation and a completion wait (agent-send). ## Core documentation - [Full AI Reference](llms-full.txt): self-contained facts, Q&A, and failure-mode catalog for this plugin - [README](README.md): human-facing narrative (English) - [README 日本語](README.ja.md): human-facing narrative (Japanese) - [CHANGELOG](CHANGELOG.md): release history ## Skills - [spawn-session](skills/spawn-session/SKILL.md): start a named, detached Claude Code Remote Control session for any project directory in its own Herdr pane, with optional model, effort, permission mode and first prompt; includes spawn.sh - [agent-send](skills/agent-send/SKILL.md): send a prompt to an agent in a Herdr pane and report whether it landed, or wait until the work is done; includes agent-send.sh ## Plugin manifests - [plugin.json](.claude-plugin/plugin.json): Claude Code plugin manifest (v2.1.1, MIT) - [marketplace.json](.claude-plugin/marketplace.json): self-owned marketplace pointing at this repo ## Design records - [2.1.0 plan](docs/plans/herdr-toolkit-2-1.html): the plan behind agent-send and the spawn-session changes - [Session bottlenecks](docs/plans/research/2026-10-10-herdr-session-bottlenecks.md): counts from 1,704 herdr calls in the author's sessions - [External research](docs/plans/research/2026-10-10-herdr-toolkit-next.md): Herdr 0.9.3, Claude Code 2.1.296 and plugin facts as of 2026-10-10 ## Related - [Herdr](https://github.com/herdrdev/herdr): the upstream terminal agent multiplexer (Apache-2.0, by ogulcancelik); prints its own herdr CLI skill with `herdr --skill` - [claude-harness](https://github.com/shimo4228/claude-harness): the author's aggregate harness export, where these skills also appear - [akc-cycle](https://github.com/shimo4228/akc-cycle): the author's Claude Code plugin that installs the Agent Knowledge Cycle (audit and upkeep skills plus an optional rules file) - [harness-scope](https://github.com/shimo4228/harness-scope): the author's Claude Code Mod that turns global skills, agents, rules and tools on or off per repo with named profiles - [shimo4228](https://github.com/shimo4228/shimo4228): the author's hub, with the long-running practice lines and their DOIs
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.