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/iwe-org/dev-workspace/shipnpx skills add iwe-org/dev-workspace --skill shipgit clone --depth 1 https://github.com/iwe-org/dev-workspaceWhat 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 | $0.00079 | $0.00865 |
| Opus 5 | $0.00039 | $0.00432 |
| Sonnet 5 | $0.00016 | $0.00173 |
| Haiku 4.5 | $0.00008 | $0.00086 |
Grade A, and why
ship 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 2d ago.
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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ship finished work
Shipping is a state change across the whole graph, not one frontmatter flip. Specs sync first — a plan is not done while the specs it invalidated still describe the old behavior.
Steps
- Confirm. Identify the plan and check its
## Verificationsection against reality (test output, the diff, a manual check) — the verify skill (.claude/skills/verify/SKILL.md) is the thorough form of this step, and a clean verify is the best possible input to shipping. If verification can't be confirmed, stop and say what's missing — don't mark done on hope. - Sync specs. Walk the plan's
## Spec changeslist. For each entry, update thedata/spec/doc to describe the behavior that actually shipped (create the doc if it doesn't exist yet, link it fromdata/spec.md, and replace the back-ticked name in the plan with a real link). Requirement / Scenario format; update scenarios that the change falsified. - Flip the plan.
iwe update -k data/plans/<key> --set stage=done --set completed=<today>, then move its link indata/plans.mdfrom## Activeto## Done. (Abandoned instead of shipped?--set stage=cancelled, move to## Cancelled, and stop here.) - Update the feature. Set the feature doc to
stage: implemented(create it from the plan's Context if it doesn't exist, linked fromdata/features.md). Fixed a bug instead? Set the bug doc tostage: done. - Record in the release. Add an inclusion link to the feature doc under
## Addedindata/releases/unreleased.md(bug fixes under## Fixed, linking the bug doc). - Cut a release (only when asked, e.g. "cut release 0.2.0"):
iwe rename data/releases/unreleased data/releases/<X.Y.Z>iwe update -k data/releases/<X.Y.Z> --set version=<X.Y.Z> --set date=<today> --set stage=released- Recreate
data/releases/unreleased.mdfresh (version/stageunreleased, empty## Added/## Fixed) - In
data/releases.md, keepUnreleasedon top and add the new version right below it (newest first).
- Log it. Append a bullet to
data/log.mdunder a## YYYY-MM-DDgroup for today (create the group if it isn't there, newest group first), one line per state change, each linking the document it describes —- **Update**: [Focus sessions](features/focus-sessions.md) implemented.A release cut gets its own line linking the release page. - Validate & commit.
iwe normalize, theniwe schema validate— must pass. Commit with a message describing the state change, e.g.ship: focus-sessions done, timer spec synced, 0.2.0 cut.
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.
- 2d ago First seen · 60 lines · 79 tokens per session scan A 96cf3e3ec48b
ship is a skill published in the GitHub repository iwe-org/dev-workspace (5 stars, last pushed 23d ago), licensed MIT. It adds 79 tokens to every session and 865 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 2026-08-31.
Other skills, from other repositories
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
gh-credit-harvest
Harvest one community PR into a release branch with authorship and credit preserved, verified green, and a warm thank-you.
pptx
从论文、大纲或结构化文本生成 PowerPoint (.pptx) 演示文稿。Use when 用户需要把一篇论文/文章/大纲做成幻灯片、slides、演示文稿、PPT、deck。Don't use when 只需纯文本总结、生成 Word/PDF、或修改已有 pptx 的单个像素级样式。.
ai-style
当任务是用中文撰写或改写面向读者的文案（产品发布稿、公众号文章、邮件、README 等）， 或用户反馈文字「AI 味太重」「不像人写的」时，加载本 Skill。.
curly-quote-sft
Skill "curly-quote-sft" from bojieli/ai-agent-book, covering 中文技术文档符号与引用规范, 何时加载, 符号定义, 决策优先级 and 正反例约束.
triage
你是当前任务的分诊协调者。先识别用户的全部目标、顺序依赖和验收条件，再按 “事实检索 → 计算/执行 → 写作”顺序逐步请求切换到需要的专业能力。不要替专业 能力完成它的工作，也不要在信息缺失时臆造结果。.