Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/RainFlashPoint/capital-agent-skillsnpx agentmods add skills/rainflashpoint/capital-agent-skills/cap-releaseWrote 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/rainflashpoint/capital-agent-skills/cap-release)<a href="https://agentmods.dev/skills/rainflashpoint/capital-agent-skills/cap-release"><img src="https://agentmods.dev/badge/skills/rainflashpoint/capital-agent-skills/cap-release.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 | $0.00319 | $0.03487 |
| Opus 5 | $0.00160 | $0.01743 |
| Sonnet 5 | $0.00064 | $0.00697 |
| Haiku 4.5 | $0.00032 | $0.00349 |
Grade A, and why
cap-release 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 3d 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 — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cap-release — 发布:目标选择器 + 晋级门引擎
入口先执行
../cap/SKILL.md的“研发上下文门禁”：.cap检查不可跳过，知识库注入失败可记录后离线继续。
全程契约：开始实质工作前读取并执行
../cap-flow/references/progress-protocol.md与../cap-flow/references/task-reconnaissance.md。先播报当前动作和下一步；新任务没有新鲜.cap/task-context.md时，先调查当前仓库代码，不能只依赖 PROFILE。
你正在执行研发主线的发布阶段。显式 local / local-only 仓使用与当前 HEAD 一致的本地 cap-gate；团队模式必须由 Server canonical state 证明同一 Commit 的 Review Action 已通过。现在由你把它逐级晋级
安全送上线。两条铁律贯穿全程:
- 每升一级先过门,过不了就回滚——deploy 完立刻 smoke/health,不健康就切回上一个好版本。
- 绝不跳级——没过 staging 不碰 canary,没过 canary 不碰 full;晋级到 full 前必须人工显式确认。
本阶段的收薄哲学:SKILL.md 只做 目标选择 + 门引擎;每种部署目标"具体命令长什么样"下沉到
targets/<type>.md(static / container / vps)。换目标类型 = 换一张卡,门引擎不变。 引擎 = 一个能 Read/Edit/Bash/Grep 的模型 + 目标仓已有的部署工具(vercel / docker / kubectl / ssh…)。
三层知识分工(别混):
- ① 门引擎(本文) —— 晋级链的决策:何时 deploy、何时探活、过门 / 回滚。平台无关。
- ② 目标适配器(
targets/*.md) —— 目标类型的命令骨架 + 回滚骨架。按目标类型分,不按语言分。 - ③ 项目特定值 —— 运行时从目标仓抽(项目名 / 集群 / 主机 / 域名 / env 清单),绝不预置在卡里。
0. 可移植前置(入口先做)
共享 references(如
deployment-patterns.md、languages/*)物理在cap-flow/references/下; 目标适配器targets/*.md就在本阶段目录(cap-release/targets/)下。解析路径分别指向两处,别混。
0.1 交互降级 —— 纯文本编号选项
发布是高风险动作,绝不自动连环上线。每次要晋级到下一环境前,用纯文本编号列表让用户确认:
staging 已部署 + smoke 通过。晋级到 canary(线上小流量)?
1) 晋级(推荐)
2) 暂停在 staging
3) 回滚 staging
回复编号即可。
宿主有结构化提问控件时可用,但回退路径必须是上面这种编号文本。晋级到 full(全量)前必须显式确认。
0.2 并行降级 —— 发布本质串行
环境逐级晋级有强依赖,且不能两处同时改线上状态。只读取证(拉日志 / 查指标)可并行;真正的 deploy / promote / rollback 一律串行 inline,一次只推一处。
1. 入口条件(进来前必须成立)
| 条件 | 检查 | 不满足 |
|---|---|---|
| review 已 PASS | local/local-only：STATE 有 cap-gate: PASS reviewed-head=<当前 HEAD>；team/server：Server currentCommit=HEAD 且 Review Action succeeded |
回 cap-review(没可信评审证据不发布) |
| verify 已收口 | STATE.stage 已过 review / verify 通过 | 先完成 verify |
| 改动已提交 | 工作树干净、HEAD 即要发布的版本 | 先提交 |
| 知道往哪发 | 能从 PROFILE.Deploy 或目标仓读到部署目标类型 |
走 §2 探测;探不到 → 编号文本问用户 |
本地
cap-gate或 Server Action 的 source commit 与 HEAD 不符(评审后又改了码)→ 停,回 review 重审。团队模式的受保护分支不接受 STATE 文本自证，由 Server Gate 负责晋级。
What ships with it
3 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.
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.
- 3d ago Changed · +2 lines 10191cdda344
- 7d ago First seen · 194 lines · 319 tokens per session scan A c491cc023fc8
cap-release is a skill published in the GitHub repository RainFlashPoint/capital-agent-skills (2 stars, last pushed 3d ago), licensed MIT. It adds 319 tokens to every session and 3,487 once invoked, about $0.0016 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
release-announcement
Write a release announcement — changelog, blog post, in-app note, or social post — that leads with user impact, names the audience, and includes upgrade/migration steps without filler.
agent-release-swarm
Agent skill for release-swarm - invoke with $agent-release-swarm.
multi-agent-release-manager
Cleans up the workspace, formats code, runs presubmit checks, and uploads CLs to Gerrit.
release-notes
Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.
pack-submit
Package one of this agent's own skills as a standalone community pack and submit it to the aeon registry as a PR.
updater_guide
Guidance for checking for and installing Row-Bot updates.