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/ccdawn/vibelution/vibelution-continuous-optimization-loopnpx skills add CCDawn/Vibelution --skill vibelution-continuous-optimization-loopgit clone --depth 1 https://github.com/CCDawn/VibelutionWrote 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/ccdawn/vibelution/vibelution-continuous-optimization-loop)<a href="https://agentmods.dev/skills/ccdawn/vibelution/vibelution-continuous-optimization-loop"><img src="https://agentmods.dev/badge/skills/ccdawn/vibelution/vibelution-continuous-optimization-loop.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 5.1 | $0.00065 | $0.00820 |
| Opus 5 | $0.00032 | $0.00410 |
| Sonnet 5 | $0.00013 | $0.00164 |
| Haiku 4.5 | $0.00006 | $0.00082 |
Grade A, and why
vibelution-continuous-optimization-loop 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 6d 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.
What it actually says
Vibelution 持续优化 Loop
每轮 = 一次 ROI 迭代。禁止一轮多 item、禁止无验证宣称完成。
启动
- 固定节奏：
/loop 30m+ 本 skill - 动态：
/loop+ 本 skill（自设 heartbeat；有 CI/log watcher 时优先事件唤醒） - 读
.runtime/loop/STATE.md（无则创建）与docs/guides/agent-dev-roi-backlog.md
红线（摘要）
- 根
main只读；写入在.worktrees/<task-slug>+codex/<task-slug> AGENTS.md→docs/guides/loop.md→docs/standards/- FE 可见 UI 必须 VUI；改
web/合入前npx tsc -b --pretty false - 无授权 push/PR；无可见控制台；合入门通过后主动
git merge --ff-only+ cleanup
单轮顺序
- 选 1 项 — backlog 最高 ROI 且
Status=todo（P0→P1→P2）；遵守热文件冲突表 - CLASSIFY — FAST_PATCH | STANDARD_TASK | HIGH_RISK
- LOCATE —
route.md+ownership.md+ 域 README + 测试 - ISOLATE — STANDARD+ 开 worktree；多 Agent 时 claim
- IMPLEMENT — 最小 diff，只改 owner surface
- VERIFY —
select_tests.py --from-git main --commands-only→ 聚焦 pytest；FE 触面 + contract +tsc -b - INTEGRATE — 自审 → ff-only merge → cleanup
- RECORD — 更新 backlog Status；append
.runtime/loop/STATE.md
机器验证（backpressure）
Agent 不得自评通过。fail → ITERATE（同 ROI，最多 2 轮）→ 仍 fail → BLOCKED。
每轮输出
Loop 轮次: <N>
ROI-ID: Rxx
决策: ADVANCE | ITERATE | BLOCKED | STOP
本轮结果: …
新证据: <命令 + pass/fail>
Agent 便利变化: …
merge: merged | not merged + 原因
下一动作: …
STOP
满足任一时末尾输出 exact 字符串：
<promise>VIBELUTION_LOOP_STOP</promise>
- P0/P1/P2 无剩余 todo（或仅 blocked 且无新证据）
- 连续 3 轮 ITERATE/BLOCKED 无新证据
- HIGH_RISK 缺用户确认
- claim/diff 冲突无法协商
禁止
- 一轮多个 ROI
- 为 loop 写长文档而不改代码/门禁
- loopmaxxing（重复已通过项）
STATE 模板
参考
- 执行环：
docs/guides/loop.md - 工作队列：
docs/guides/agent-dev-roi-backlog.md - Ralph / Loop Engineering：固定 prompt + 磁盘状态 + 独立 verifier + max iterations
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.
- 6d ago First seen · 81 lines · 65 tokens per session scan A fadf2d7e71bd
vibelution-continuous-optimization-loop is a skill published in the GitHub repository CCDawn/Vibelution (21 stars, last pushed yesterday), licensed MIT. It adds 65 tokens to every session and 820 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
background-watch-hook
Use vibe watch to run a managed Harness waiter that returns to the same conversation later. Best for reviews, CI, files, logs, and other wait-now-continue-later workflows.
google-workspace-setup
One-time setup for gws: install, OAuth, scopes, auto-approve.
google-calendar
Google Calendar via gws: list events, create, accept, find free time.
attachments
Move bytes between Gini upload space, external URLs, and workspace files. Used by every attachment / file-upload / file-download flow regardless of the target system (Linear, GitHub, S3, Notion, etc.).
gini-bug-report
File a locally-captured, already-redacted Gini crash report as a GitHub issue, with the user's consent. Reads the pending crash queue and delegates the actual filing to the github-issues skill.
google-docs
Google Docs via gws: read, append text, structured batch edits.