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.
npx skills add bahayonghang/my-ai-cli-toolkit --skill herdr-orchestragit clone --depth 1 https://github.com/bahayonghang/my-ai-cli-toolkitWrote 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.
[](https://agentmods.dev/skills/bahayonghang/my-ai-cli-toolkit/herdr-orchestra)<a href="https://agentmods.dev/skills/bahayonghang/my-ai-cli-toolkit/herdr-orchestra"><img src="https://agentmods.dev/badge/skills/bahayonghang/my-ai-cli-toolkit/herdr-orchestra/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.
<a href="https://agentmods.dev/skills/bahayonghang/my-ai-cli-toolkit/herdr-orchestra"><img src="https://agentmods.dev/badge/skills/bahayonghang/my-ai-cli-toolkit/herdr-orchestra.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00084 | $0.00896 |
| Opus 5 | $0.00042 | $0.00448 |
| Sonnet 5 | $0.00017 | $0.00179 |
| Haiku 4.5 | $0.00008 | $0.00090 |
Grade A, and why
herdr-orchestra 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 today.
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.
What it actually says
Herdr Orchestra
在 Herdr 中把明确任务交给 worker,并向调用者交付可核验的结果。默认一项任务、一个 worker;调用者负责权限、分工和最终验收。
Router Rules
- 只接管显式 Herdr 编排请求,或已获授权的上层 skill 委派。普通 review 走既有审查流程,简单 pane 查询走官方
herdr指引或 CLI help;研究 Herdr 文档不启动进程。 - 已被分配为 leaf 的 worker 直接做本职工作,不再次调用此入口。新进程不会自动继承调用者加载的技能:把必要约束放进任务文本。
- 非 Herdr 环境停止本 skill 的会话操作,并说明限制;缺少能力时不偷偷改用其他传输。
Compact Workflow
- 阅读 委派契约,取得目标、cwd、kind/native argv、输入范围、权限/文件 ownership、输出和验收条件。未指定模型就沿用用户配置。未授权的范围变化才需要补问。
- 在任何 Herdr 会话控制前检查
HERDR_ENV=1。使用当前 CLI help 确认语法,通过 caller context 取调用 pane;不以 focused pane 代替调用者。 - 按契约在创建 sibling 与
agent start前选择 surface:用户明确要求交互 TUI/Agent 则 agent;否则 kind 原生非交互命令能满足验收则 pane;其余 agent。默认在调用 tab/cwd 创建一个 sibling,按布局选 right/down,保持--no-focus。解析返回 ID,确认 shell 可用。 - pane surface:
pane run与非碰撞 wait,完成证据是命令退出后的输出,不是 agent idle/done。agent surface:agent start指定 kind,确认 ready identity 后向任务自有且 idle 的 worker 发送一次完整 assignment,使用agent prompt --wait和有限 timeout。start 或 identity 失败时按契约检查同一 pane 并 incomplete;不重复提交或自动替换。blocked、unknown、旧输出都不能证明完成。 - 按契约收集完整响应,区分 passive read 与可能滚屏的 history read。只读 worker 的截断响应以小段重述恢复;文件 fallback 需要已有写权限。核验实际 scope、证据和缺失项。
- 交付下列 receipt,保留 worker pane。需要进一步独立分工或异常恢复时阅读 模式与恢复。提交、发布、修复、trust、更改集成或扩大清理均不因编排而获得授权。
Output Contract
- 返回解析得到的 worker name/kind/pane identity,以及本任务创建的资源和保留状态。
- 分开记录已观察的 lifecycle、实际任务/输入范围、收到的响应、覆盖与完整性、错误或待决事项。终端 idle/done 和任务成功是两个判断。
- 调用者对照输入核验结果;不以多 Agent 一致意见代替证据。响应仍缺失时明确 incomplete,不把它压缩成“无问题”。
Evidence
来源与设计取舍见 调研记录,当前评估和限制见 交付记录。本包提供指令契约,不声称本地检查证明真实 server、模型执行或新会话加载。
What ships with it
10 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.
- agents/interface.yaml 1.1 KB
- evals/evals.json 9.1 KB
- evals/trigger_cases.json 1.9 KB
- references/delegation.md 8.8 KB
- references/patterns-and-recovery.md 8.8 KB
- reports/creation-handoff.md 5.4 KB
- reports/output-eval.md 5.9 KB
- reports/prior-art-research.md 5.4 KB
- reports/skill-ir.json 6.3 KB
- reports/trigger-eval.json 7.2 KB
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.
- today Changed 8fa705010fc1
- 2d ago First seen · 37 lines · 84 tokens per session scan A eb7b05b7bb04
herdr-orchestra is a skill published in the GitHub repository bahayonghang/my-ai-cli-toolkit (16 stars, last pushed today), licensed MIT. It adds 84 tokens to every session and 896 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 2026-09-09.
Other skills, from other repositories
agent-creation-skill
Procedural guide for creating new agents from templates when project complexity requires delegation.
swarmclaw
AI agent runtime and multi-agent orchestration platform. Teaches agents how to use SwarmClaw's 6 primitive tools, persistent memory, dreaming, delegation, connectors, credentials, and the skill system. Use when an agent is running on SwarmClaw and needs to understand the platform's capabilities.
orchestration
Multi-agent orchestration authority — agent memory protocols, task delegation, SOUL/AGENTS patterns, session startup, inter-agent communication, OpenClaw workspace coordination, and agent lifecycle management.
pm-delegation-patterns
Common delegation patterns for PM agent.
agent-creation-skill
Procedural guide for creating new agents from templates when project complexity requires delegation.
agent-framework-py-release
Use when cutting a Python release for the microsoft/agent-framework monorepo. Triggers on "bump py versions", "cut a python release", "prepare release PR for python", "release py packages", "bump python to X.Y.Z", or similar requests to bump Python package versions and prepare a release PR. Handles all four lifecycle…