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 agentmods add skills/sd0xdev/sd0x-harness/orchestratenpx skills add sd0xdev/sd0x-harness --skill orchestrategit clone --depth 1 https://github.com/sd0xdev/sd0x-harnessWrote 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/sd0xdev/sd0x-harness/orchestrate)<a href="https://agentmods.dev/skills/sd0xdev/sd0x-harness/orchestrate"><img src="https://agentmods.dev/badge/skills/sd0xdev/sd0x-harness/orchestrate.svg" alt="Measured on agentmods" 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 | $0.00107 | $0.04277 |
| Opus 5 | $0.00053 | $0.02139 |
| Sonnet 5 | $0.00021 | $0.00855 |
| Haiku 4.5 | $0.00011 | $0.00428 |
Grade A, and why
orchestrate 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrate Skill（v1 report-only）
宣告式意圖 → planner agent 推導隨 repo 狀態變動的 workflow 計畫 → 預覽人核 → read-only fanout → pre/post 無變更驗證 → 報告走既有 doc review loop。Control plane（.claude_workflows/）是 orchestrator 唯一的狀態面；review 提醒狀態自 hook-lightweighting § 3.2 起存於 repo 外（~/.cache/sd0x-dev-flow/），orchestrator 不讀不寫。
Trigger
- Keywords: orchestrate, workflow orchestration, 編排, plan this workflow, audit the repo for, 自動編排
When NOT to Use
- 執行會改檔的步驟（v1 一律輸出為
proposed-manual，交人走正常流程） - 單一 skill 就能完成的任務（直接呼叫該 skill）
- Code review（
/codex-review-fast）、doc review（/codex-review-doc） - 既有固定形狀的深度探索（
/deep-explore）或研究（/deep-research）——orchestrate 是跨 skill 的編排層，不取代它們
Flags
| Flag | 行為 |
|---|---|
（無）/ --dry-run |
規劃 + 預覽即止（plan preview = 預設交付物） |
--execute |
預覽 + AskUserQuestion 核可後執行 read-only fanout → verify → 報告 |
--budget S|M|L |
Budget tier（default M）；超量 fail-closed，見 scripts/plan-context.js |
（內部）plan-context.js --out |
封包一律寫檔、只回摘要（含 sha256）；主 session 不得持有完整封包（~81 KB × 2 → ~0.5 KB），見 references/planner-prompt.md |
--resume <run-id> |
以原 baseline compare：無 drift → 續跑 pending 步驟；有 drift → run 標 needs_human 並停（不得重拍 baseline 洗白；要繼續只能開新 run = 新 baseline + 新核可）。baseline 本體留在 .claude_workflows/<run-id>/baseline.json，跨 session 仍在；但 baseline_sha256 只活在主 session 的 context，run-state 刻意不存它（存在 baseline 旁邊等於沒綁定）。因此 resume 的實際限制是 digest 而非檔案：脈絡遺失／換 session → 不可 resume，開新 run |
--backend dw|agent |
強制 fanout 後端；預設 auto（Workflow 工具可用則用，否則 background Agent ≤3 並行）——admission 同一套 |
Workflow
sequenceDiagram
participant U as User
participant C as Claude（主 session）
participant S as scripts
participant P as Planner（Explore，admission 約束）
participant W as Fanout workers（read-only）
U->>C: /orchestrate "意圖" --execute
C->>S: plan-context.js --out（封包寫檔，主 session 只收摘要）
C->>S: run-verify.js snapshot > .claude_workflows/<run-id>/baseline.json（digest 由 stderr 收進 session）
C->>P: 背景派發 planner（傳封包路徑，非內容）
P->>P: Read 封包
P-->>C: plan JSON
C->>S: run-verify.js compare --baseline-sha256（planning 後立即驗）
C->>S: validate-plan.js --context-sha256 <摘要 sha256>（admission + gate lint + 封包完整性）
C->>U: Plan preview + AskUserQuestion
U-->>C: 核可
C->>C: Write run-state（redact 先行）
loop 每個 parallel_group（≤ max_waves）
C->>W: 平行派發（單訊息）
W-->>C: findings（context packet）
end
C->>S: run-verify.js compare --baseline-sha256（同一 baseline，同一 digest）
alt drift
C->>U: ⚠️ Need Human（run 標 failed，不寫報告）
else 無變更
C->>C: Write 報告 → /codex-review-doc 接手
end
What ships with it
8 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.
- references/admission-allowlist.json 4.2 KB
- references/execution-policy.md 2.2 KB
- references/plan-schema.md 3.2 KB
- references/planner-prompt.md 6.9 KB
- scripts/plan-context.js 22 KB runs code
- scripts/prune-runs.js 21 KB runs code
- scripts/run-verify.js 65 KB runs code
- scripts/validate-plan.js 22 KB runs code
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.
- yesterday First seen · 132 lines · 107 tokens per session scan A ed7cc5e99030
orchestrate is a skill published in the GitHub repository sd0xdev/sd0x-harness (188 stars, last pushed yesterday), licensed MIT. It adds 107 tokens to every session and 4,277 once invoked, about $0.0005 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 2026-09-03.
Other skills, from other repositories
cc10x-router
THE ONLY ENTRY POINT FOR CC10X. Activate this skill for build, debug, review, and plan requests. Use when the user asks to implement, fix, review, plan, test, refactor, or continue code work. Trigger keywords: build, implement, create, write, add, review, audit, debug, fix, error, bug, broken, plan, design, architect…
building
Implementation skill for writing production code with TDD. Covers the RED-GREEN-REFACTOR cycle, false-RED detection, vertical slicing, scope escalation, test process discipline, and code generation patterns. Loaded by component-builder and bug-investigator.
diff-driven-docs
Use when a BUILD phase completes, a commit is staged, or a PR is about to be created, and the diff has not yet been reflected in documentation. Also use when the user says "update docs", "sync docs", "document this", or asks whether documentation is up to date.
exploration
Two-mode exploration skill: (1) design dialogue — turn rough ideas into validated designs through collaborative interview before planning; (2) spike — throwaway code answering ONE design question, deleted or absorbed, never shipped. Router invokes mode via dispatch context.
memory-and-handoff
Two-mode skill: (1) session memory — load/persist durable workflow state under .cc10x/ (activeContext, patterns, progress) so context survives compaction; (2) handoff package — portable, secrets-redacted export for a coworker, different tool, or fresh non-cc10x session.
plan-review-gate
Use after saving a non-trivial plan or decision RFC when a fail-closed feasibility, completeness, and alignment review must block execution.