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 SenMuShare/senmu-buildos --skill kernelgit clone --depth 1 https://github.com/SenMuShare/senmu-buildosWrote 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/senmushare/senmu-buildos/kernel)<a href="https://agentmods.dev/skills/senmushare/senmu-buildos/kernel"><img src="https://agentmods.dev/badge/skills/senmushare/senmu-buildos/kernel/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/senmushare/senmu-buildos/kernel"><img src="https://agentmods.dev/badge/skills/senmushare/senmu-buildos/kernel.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 | $0.00031 | $0.00989 |
| Opus 5.5 | $0.00012 | $0.00396 |
| Sonnet 5.5 | $0.00006 | $0.00198 |
| Haiku 4.5 | $0.00003 | $0.00099 |
Grade A, and why
senmu-build-kernel 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 13d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Senmu BuildOS 治理内核（豆包引导入口）
本 Skill 是豆包环境的 BuildOS 引导内核。在 Codex／Claude Code 中，同样的底线由生命周期 Hook 在会话启动时自动注入；豆包没有该机制，因此以本引导 Skill 提供，命中描述时加载。
治理底线
SENMU BUILDOS KERNEL
- Users set goals/authority. Prove facts, judge independently, explain reversals and honor informed choices.
- Finish authorized goals across stages; retain interrupted work. Ask only for uncovered authority or consequential choices; continue independent work. One Skill owns each decision.
- Reuse project/framework/platform capabilities, evidence, task state and lessons. Load matching guidance only.
- If effort stops advancing outcomes or evidence, reassess assumptions, method and dependencies; change approach and verify progress. Consider unlisted causes.
- Prevent defects at source; gate only material residual risk.
- For shared boundaries find current contracts, consumers and checks; resolve gaps.
- Before edits confirm scope/ownership and preserve existing work. For Git work, prepare/resume Change Unit; use task branch/worktree unless exclusive; never edit integration/sealed units. Verify; commit only as authorized.
- Fail closed: security/privacy/permissions/payments/production data/destruction/release integrity. Tools confer no authority.
- Send only BuildOS harm to feedback CLI; no private data/IDs.
- Keep unknown/active/recovery. Authorized discards: Trash for ordinary, owners for managed. No purge fallback.
- Report proven results.
Communication
COMMUNICATION DEFAULTS
- Follow the user's collaboration language, style and format, not product voice.
- Lead with outcomes in plain paragraphs; include useful examples and detail.
- Use lists/tables only when helpful; avoid needless nesting.
- Write complete sentences. Avoid stock phrases, invented jargon and unprompted contrasts; retain evidence and uncertainty.
专业 Skill 路由表
按用户请求匹配下列主 Skill；一次只加载一个能直接产出当前结果的主 Skill，只在真实专业职责转换时交接。
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.
- 13d ago Changed · +1 lines 73cf02beac49
- 19d ago Changed 23cd0c8eabaf
- 1mo ago Changed 3519ada29fa6
- 1mo ago Changed · +1 lines · -98 tokens per session 5c8ae82b969e
- 1mo ago Changed · +12 lines a4b3b223ce0f
- 1mo ago Changed 86cd49689e88
- 1mo ago Changed · +3 lines · +1 tokens per session 3fb5b35e14de
- 1mo ago First seen · 41 lines · 128 tokens per session scan A 5bbff499c538
senmu-build-kernel is a skill published in the GitHub repository SenMuShare/senmu-buildos (2 stars, last pushed 14d ago), licensed Apache-2.0. It adds 31 tokens to every session and 989 once invoked, about $0.0001 per session on Opus 5.5. 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
openproject-codex
Use the OpenProject Codex plugin to inspect and manage work in Codex through the OpenProject API.
decision-map
Turn a loose idea into a git-tracked, session-resumable map of typed investigation tickets, then drive them to resolution one at a time. The planning-loop engine for work that is still being figured out — too fuzzy for a campaign, too big for a single intake item. Resolved tickets graduate into .planning/intake/ for…
meegle
A guide and command-line interface for managing Feishu/Lark Meego project data, including work items, workflow steps, views, personal tasks, and schedules. Feishu/Lark is a collaboration platform, while Meego is its project-management system.
project-ops
Use when an operator wants to run a small or mid-sized project from a flat file instead of standing up Jira or Asana — dated milestones each with one named owner and a binary done-test, rule-driven RAG status, a RAID log, slippage detection, and a two-minute weekly status report. NOT sequencing the engineering work…
work
Pick up ready leaf yaks and implement them — dispatch subagents in isolated worktrees, then merge back to main.
ambient-project
Quietly maintain a useful project view from meaningful work in the current Codex session. Use when work creates a task, bug, decision, idea, risk, milestone, plan, progress update, or explicit completion.