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/stefan-stepzero/shipkit/shipkit-orchestratenpx skills add stefan-stepzero/shipkit --skill shipkit-orchestrategit clone --depth 1 https://github.com/stefan-stepzero/shipkitWhat 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 | $0.00102 | $0.02232 |
| Opus 5 | $0.00051 | $0.01116 |
| Sonnet 5 | $0.00020 | $0.00446 |
| Haiku 4.5 | $0.00010 | $0.00223 |
Grade A, and why
shipkit-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 — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
shipkit-orchestrate — Core Automation Engine
Purpose: One reusable orchestration loop that every phase skill calls, so Shipkit has exactly one way to drive work to completion instead of a separate hand-rolled loop per phase. It dispatches units, reconciles their output against ground truth (never self-reports), re-dispatches the gaps, integrates, and stops at an agreed bar.
Read first: install/shared/references/core-automation.md — the contract (substrate
choice, the worktree finding, partition, the ledger, ground-truth reconciliation,
integration policy, the parallel-build lifecycle, deterministic-mechanism-vs-judgment).
This SKILL.md is the entry point; the contract is the machinery.
Runs inline in the main session. The engine launches Workflow fan-outs and drives
git + AskUserQuestion, all of which require the main session — it is not a forked
orchestrator. A phase skill that needs parallel fan-out hands its plan up to the
engine (which the main session is running), and the engine launches.
Calling interface
A caller invokes the engine with four things:
| Input | What it is |
|---|---|
| steps | the ordered / parallel units to drive — each a { skill or work, artifact } pair (what to run, what it must produce) |
| bar | the ground-truth completion criteria per step — the named .shipkit/ artifacts that must exist + any check (tests green, review passed). This is what "done" means; the caller must state it. |
| mode | autonomous or steered (see references/modes.md) |
| elicitive | which steps may need user input — handled via the marker-bubble, never inside a workflow |
The engine owns the loop; the caller owns the WHAT. No caller re-implements dispatch / reconcile / re-dispatch — that lives here, once.
Modes (governance)
See references/modes.md for the full treatment. In brief:
- autonomous — run the steps to the bar unattended; reconcile against ground truth each round; re-dispatch gaps; stop at the bar or the slot ceiling. Used by build and direction-propose. (Shipkit-native drive-to-done.)
- steered — the main session coordinates; one unit per area; surfaces back to the user to steer between rounds. Used by review and interactive build. (Shipkit-native team-lead.)
- scale toggle — a resilience flag on autonomous (rate-limit-aware pacing) for very large fan-outs; not a separate mode.
What ships with it
1 file 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.
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 · 189 lines · 102 tokens per session scan A e322c302b8ef
shipkit-orchestrate is a skill published in the GitHub repository stefan-stepzero/shipkit (1 stars, last pushed 1mo ago), licensed MIT. It adds 102 tokens to every session and 2,232 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 2026-08-31.
Other skills, from other repositories
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
golden-rss
Use when testing the rss golden build.
omh-code-review
This is a Hermes-native code-review workflow skill.
redteam-web-detail-pack
Routing and boundary guidance for authorized general web application security testing. Use as a web testing router when the attack surface should be dispatched to more specific web vulnerability skills.
android-pentest
安卓应用渗透测试 — APK分析、Hook、自动化测试、运行态驱动、签名恢复、抓包分析.
studio
Architecture Studio control plane — initialize or inspect a studio workspace, create and register projects, or route an architecture/AEC task to the right agent or skill. Use when the user runs /as:studio, asks to set up or open their studio, manage its projects, or describes a task without naming a skill.