Release Tag logo

Release Tag

OrganizationPopular
Prismer-AI
release-tag

Cut a release tag behind a real tier gate and cloud approval readback. The approval, task, approved commit, and resulting tag are bound; self-approval and HEAD drift fail closed. Prod prefixes remain refused.

Overview

PublisherPrismer-AI
RepositoryPrismerCloud
Skill namerelease-tag
Stars
1.6K
Forks
17
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Prismer-AI on GitHub. Read the source before you install it.

Installation

Install the Release Tag AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Prismer-AI/PrismerCloud/tree/main/sdk/apc/skills/release-tag
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Prismer-AI/PrismerCloud.git /tmp/PrismerCloud
mkdir -p .claude/skills
cp -r /tmp/PrismerCloud/sdk/apc/skills/release-tag .claude/skills/release-tag
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Release Tag in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Release Tag on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Release Tag is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

release-tag

发版打 tag(apc/01 §2 release-tag,收口步 6)。它把测试→上线的焊点焊死:没有 tier 全绿 / baseline 无新红的结构化证据,审批请求根本不创建、tag 绝不 push。tier 门跑真 run.ts 读真退出码;tag 名按 <channel>-<target>-YYYYMMDD-vX.Y.Z 算;审批通过后 git tag && push 到本地 bare mirror 替身(真 GitLab push + pipeline 轮询 = M5)。

什么时候用:release-preflight 绿 + release-db-config-sync 无阻塞之后,要把一个 test 版本推上 CI。

承重纪律(发版 skill 的命根子,先记死)

  • prod 前缀 tag 必拒(不变量 2 · 两人闸)。四个映射 APP_ENV=prod 的前缀——k8s-prod-* / desktop-prod-* / prod-* / ali-k8s-prod-*——先于一切 git / tier 动作硬拒,本 verb 不做任何 git 动作。别试 prod:真 prod 红线由 GitLab protected tags(P-1,运维配置)落地,本 skill 只护到"不替你打 prod tag"。要发 prod = 人走 M5 路径。
  • --tier 必须显式传,没有默认值。TD(桌面冒烟)在 apps/desktop/e2e/smoke/ 无 spec 时会自动 skipped-with-note(passed:0/failed:0),单独拿它当发版门是一道不可能红的空门——之前 SKILL 范例全不带 --tier 让 agent 照抄默认值正是这么踩的坑(apc/12 §0.12)。选一个会真执行断言的层,例如 --tier=T0,T1;prod 前缀路径(下方触发)例外——它先于 tier 门拒绝,不需要 --tier。
  • tier 绿是 push 的硬前置,"绿"必须是真跑过用例。tier 红 / env_blocked / 空跑(该轮所有 tier 的 passed+failed 合计为 0,即一个用例都没真执行)⇒ blocked,审批请求不创建、tag 不 push。不许放松 tier、挑能绿的层、或拿一个恰好全 skip 的层骗过门。
  • 审批门不可跳。tier 绿但未审批 = staged(dry-run:出 tag + tier 证据,不 push)。批准路径必须同时传 --task <taskId> --approved <approvalId>;CLI 从 cloud 读回批件和 task,校验 approved / release_tag / task 绑定 / 非自批,且两者 metadata 的 commit SHA 必须一致。随机 token 不再能放行。
  • tag 打获批 commit,不打 HEAD。并行工作使 HEAD 漂移也不得把未审内容带入 tag;获批 SHA 不在 repo 则 fail-closed。

Anti-pattern(直接判红)

  • ❌ 给 --target prod 或手拼 prod 前缀 tag 想让 skill 打 prod。命中即 blocked,别绕。
  • ❌ tier blocked 时先打 tag 再补测试。焊点就是不让这发生。
  • ❌ 拿到 staged 就以为发版完成。staged 是"等审批",tag 还没 push。
  • ❌ 用 git tag + 裸 git push origin 手工绕过本 verb,跳过 tier 门和审批门。

工具契约(签名以此为准)

apc 不在 PATH——在仓库根用 npx tsx sdk/apc/bin/apc.ts <sub>。

命令作用退出码
npx tsx sdk/apc/bin/apc.ts release tag [--channel k8s|desktop] [--target test|prod] [--version X.Y.Z] --tier=<tiers> [--task <taskId> --approved <approvalId>] [--json]prod 人闸 → tier 门 → cloud 批件/task/commit 他证 → 把获批 SHA push 到本地 bare mirror0 green(pushed)· 3 staged(tier 绿待审批)· 1 blocked(tier 红 / 他证失败 / prod 拒 / push 失败)· 2 用法错

--tier 没有默认值(--target prod 路径除外——它先于 tier 门拒绝);不传直接 usage 错 exit 2。选一个会真执行断言的层,例如 --tier=T0,T1——别用 --tier=TD 当唯一层:无桌面冒烟 spec 时它是空门。

--json 字段(名以此为准):

  • 顶层:{ verb:'tag', decision, tag, target, channel, tier, approved, approvalId, taskId, commitSha, pushed, mirror, blockers:[], notes:[] }
  • tag:算出的 tag 名(如 k8s-test-20260724-v2.0.7);--version 缺省读 /VERSION
  • tier:{ tier, exitCode, envBlocked, regressions, emptyRun }(prod 拒时 exitCode:-1,因为压根没跑到 tier;emptyRun:true = 该轮所有 tier 的 passed+failed 合计为 0,一个用例都没真跑,即便 exitCode 是 0 也判 blocked)
  • mirror:push 目标(默认 .dev-stack/apc-bare-origin.git bare 替身)

Workflow

1. dry-run:算 tag + 过 tier 门(不带 --approved)

bash
npx tsx sdk/apc/bin/apc.ts release tag --channel k8s --target test --tier=T0,T1 --json > /tmp/tag-dry.json; T=$?
echo "tag(dry) exit=$T"
  • decision:staged(exit 3)+ tier 绿 → 拿到 tag 名,进审批。
  • decision:blocked(exit 1)→ 读 tier/blockers;tier 红 / 空跑(tier.emptyRun:true)就停手回 bugfix,别往下走。
  • 忘了 --tier → exit 2 用法错,不是"用了默认 TD"——没有默认值。

2. 审批后 push(cloud 批件 + task 绑定)

bash
# approval 必须是 cloud 中已被另一个人批准的 release_tag 记录,并且与 task 携带同一 commit SHA
npx tsx sdk/apc/bin/apc.ts release tag --channel k8s --target test --tier=T0,T1 \
  --task "$PRISMER_TASK_ID" --approved "$APPROVAL_ID" --json > /tmp/tag.json; T=$?
echo "tag(push) exit=$T"
  • decision:green(exit 0)+ approved:true + pushed:true → tag 已落 bare mirror;必须用 git rev-parse refs/tags/<tag>^{commit} 核验结果等于产物的 commitSha。

3. prod 一律拒(验证人闸,不是要发 prod)

bash
npx tsx sdk/apc/bin/apc.ts release tag --channel k8s --target prod --json; echo "prod exit=$?"
# 预期:decision=blocked, exit 1, blockers 含 "prod 人闸",无任何 git 动作
# 注意:这条命令没带 --tier ——prod 人闸先于 tier 门开火,压根不需要它

Failure / 边界

  • push 失败(bare mirror 不存在 / git 错误)→ blocked + blockers 含真 git stderr。bare 替身由 apc env up --safe 拉起。
  • 本机替身:push 只到 本地 bare mirror,不推 GitLab;cloud approval/task/commit 他证已是真路径。真 GitLab push + pipeline 轮询仍是 M5(诚实 ⬜)。

输出契约(机器判据按这个复算,别自由发挥格式)

本 skill 的判据不是「报告里出现了 staged / prod 这些字」,而是:重解析你贴的三份 --json 产物、从 decision 反推退出码、用只读 git plumbing 回读 bare mirror 的 ref 库、用 tier 门自己写的副产物给这次运行定时(structured-criteria.ts 的 json-claim + git-claim-readback)。

先在隔离沙箱里搭演示(绝不在共享工作树打 tag/push):

bash
mkdir -p .e2e-tmp/apc-release-tag && git init -q --bare .e2e-tmp/apc-release-tag/mirror.git
mkdir -p .e2e-tmp/apc-release-tag/work && cd .e2e-tmp/apc-release-tag/work
git init -q -b main && echo demo > f.txt && git add f.txt && git commit -qm init
git remote add mirror "$REPO/.e2e-tmp/apc-release-tag/mirror.git"

--repo / --mirror 必须传绝对路径——verb 内部会把相对 --repo 再解析一次(git -C <rel> 又在 cwd=<rel> 下跑),相对路径必失败。

报告必须带下面这组行 + 三份原样 JSON:

RUN-AT: <审批那次运行的 ISO-8601 墙钟时间>
STAGED-TAG: <dry-run 产物的 tag>
STAGED-PUSHED: false
STAGED-EXIT: 3
TAG-NAME: <审批产物的 tag>
TIER-EXIT: 0
TAG-EXIT: 0
PROD-TAG: <verb 算出并拒掉的 k8s-prod-… tag>
PROD-DECISION: blocked
GIT-REPO: .e2e-tmp/apc-release-tag/work
GIT-TAG: <tag> | remote: mirror | sha: <该 tag 在 mirror 上解析出的 commit sha>
GIT-REFUSED-TAG: <k8s-prod-… tag> | remote: mirror

紧跟着把三次 --json 的 stdout 一字不改贴进三个 fenced json 块(dry-run → 审批 → prod)。

判据会判红的情况(任一):

  • 没有可解析的 fenced JSON(只有散文「tag 打好了」);
  • STAGED-EXIT / TAG-EXIT 不是从 decision 推出来的那个(green→0 · staged→3 · blocked→1)——退出码是 decision 的函数,不是措辞;
  • 审批产物 decision:green 却 tier.exitCode != 0 / tier.regressions 非空 / blockers 非空——tier 绿是 push 的硬前置;
  • dry-run 产物 approved:false 却 pushed:true——审批门被跳;
  • prod 产物不是 blocked + pushed:false + tier.exitCode:-1——-1 表示人闸先于 tier 门开火,压根没跑到测试,写成 0 就等于承认跑过 tier,判红;
  • GIT-TAG 声明的 tag 不在 mirror 的 ref 库里,或它解析出的 commit ≠ 你声明的 sha——「push 了」是 ref,不是句子。未审批的 dry-run 不许声明 GIT-TAG(它什么都没推);
  • GIT-TAG 的 tag 名不匹配 <channel>-test-YYYYMMDD-vX.Y.Z,或它本身命中 prod 前缀;
  • GIT-REFUSED-TAG 竟然在 mirror 上——那不是拒绝,是推上去了;
  • GIT-REPO 不在 .e2e-tmp/ 下,或指向打分器自己的工作树;
  • RUN-AT 与 tier 门副产物 scripts/test203/artifacts/desktop-shell-skip-evidence.txt 的 mtime 差超过 20 分钟——tier 门没真跑,磁盘上就没有这一刻的痕迹。

诚实边界:①RUN-AT 是你声明的墙钟时间,判据只能证明它与副产物 mtime 同期,不能证明它就是"现在"——它杀的是「拿几天前的旧产物冒充这次跑」,不是「重放一次一致的旧快照」。②tier 副产物路径是按本 sample task 钉的 TD 层,换 tier 必须同步换 args。③判据不证明 mirror 上那个 commit 的内容对,只证明 tag 真在、真指向你声明的 sha。

产出(副作用 oracle,报告里必须给)

  1. decision + 真退出码:green/staged/blocked ⇄ 0/3/1。
  2. 算出的 tag 名:<channel>-<target>-YYYYMMDD-vX.Y.Z 全字打印。
  3. tier 门证据:tier.exitCode + regressions(tier 红时列出)——证明"没绿不 push"。
  4. push 副作用:green 时 pushed:true + mirror 路径(git ls-remote 可核验真 tag ref)。
  5. prod 拒证据:--target prod → blocked + exit 1 + "prod 人闸" blocker + 零 git 动作(不是聊天里说"prod 不能发")。

不许:把 staged 当发版完成;试图打 prod tag;绕过 verb 手工 tag/push;tier 红仍 push。

Checklist

  • 先 dry-run 拿到 tag 名 + tier 绿证据,再带 --approved push
  • green 时确认 pushed:true + 记录 mirror;用 git ls-remote 复核 tag ref
  • 演示 prod 前缀被拒(blocked/exit 1/零 git 动作),没试图绕
  • tier 红 / env_blocked 时停手,没有放松 tier 或先 tag 后测

PKF 直写(研发回环 · doc07 §B2 写入点①)

产出时机:tag 收口(green/staged/blocked)后,把这次打 tag 的结论直写成一张 PKF 记忆页(pageType=decision,ReleaseCard 语义 apc/09 §1)——同一页既是回审批人侧的富报告,也是可召回的发版记录。经验教训另走 memory skill 的 CONSTRUCT→PLACE→WRITE,两条边并存(doc07 §B1)。这条直写是投影,不改 prod 人闸 / tier 门 / 审批门任何一步。

语法照 pkf-writing skill——本 skill 不再内嵌语法骨架(frontmatter / typed link / 数据块写法都在那边)。写入面不新造:code agent 走 prismer memory write(SS-14 §4.3 appendix;hermes 侧是 native memory_write)。

声明(doc10 §2.5 格式,报告末尾一行):

PKF: prismer://workspace/<ws>/memory/<path>

写入面不可达时如实声明 PKF: none — 写入面不可达(<哪一条>),绝不允许为了满足规则而假装写了。

回读:声明后必须回读该页(pkf_read / memory read),确认真实存在、正文非空且与结论一致;typed link 的 prismer:// 目标必须真实存在(validator 规则 2c/4 挡 invalid-prismer-host),无对应页就删该 link 行、宁缺勿造伪目标;写入自动挂 INDEX 反孤儿锚(SS-14 §4.2),优先 edit 既有页而非 dump 新叶(PLACE 治理照走)。

调用回执(APC 平台方运营纪律 · 自动注入,勿手改)

执行本 skill 的第一步,先落一条调用回执:

bash
cloud skill ack release-tag --task "$PRISMER_TASK_ID"

按退出码分流(这条命令的退出码是承重信息,禁止用 || 兜底、; true、 set +e 或重定向把它抹掉):

exit含义你要做的
0回执已落库(im_task_logs.action='skill_ack')继续执行本 skill
3无 task 上下文——本次运行没有 task,产不出回执继续执行本 skill;但本次运行没有回执,任何报告里都不得声称已 ack
4你不是该 task 的 assignee,服务端拒绝停下并上报:回执只能由执行该 task 的 agent 产生
1其它失败(网络 / 服务端)重试一次;仍失败则继续执行,并在结果里显式标注「回执缺失」

回执只证明本 skill 被调度,不证明执行正确——效果证明由本 skill 自己的 acceptanceCriteria 副作用断言承担(apc/04 §2 层 1 诚实标注)。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Release Tag AI skill do?

Cut a release tag behind a real tier gate and cloud approval readback. The approval, task, approved commit, and resulting tag are bound; self-approval and HEAD drift fail closed. Prod prefixes remain refused.

Why use Release Tag on TypingMind?

Because you install it once and use it with any model. Release Tag is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Release Tag in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Prismer-AI/PrismerCloud/tree/main/sdk/apc/skills/release-tag. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Release Tag?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Release Tag?

As many as you like. As long as a model supports skills, you can use Release Tag with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Release Tag AI skill free?

Yes. It is published on GitHub by Prismer-AI under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇