AGENT SKILL

accept

Use as a terminal node that emits accepted@1 only when an upstream verdict@1 says pass, and fails loudly on a rejected draft regardless of how routing reached it. Single tool. Reusable standalone.

License 许可
MIT
最近更新
2026-10-30
Stars
★ 211
Forks
21

01 前言 Front Matter

name
accept
description
Use as a terminal node that emits accepted@1 only when an upstream verdict@1 says pass, and fails loudly on a rejected draft regardless of how routing reached it. Single tool. Reusable standalone.

02 资源文件 1

03 SKILL.md 正文

Skill — accept(終端節點,不信任路由)

圖示範流程的終端節點。刻意不信任路由:即使編排把一份 reject 的判定送到這裡 (例如有人手改了 flow.json 的 branch 條件),這一步仍拒絕通過。 閘門查真相,不查「我是被誰叫來的」。

綁定的單一工具

  • python3(標準庫)。本 skill 只此一個工具。

依賴

  • python >= 3.8

參數化(無硬編碼)

  • 輸入:--in <verdict artifact.json>;輸出:--out <accepted artifact.json>

介面

  • 輸入 artifact:verdict@1(讀 data.verdict、data.round)。
  • 輸出 artifact:accepted@1 — data{accepted, rounds, draft_status:"DRAFT"},透傳 trace。

fail-loud 邊界

data.verdict != "pass" → exit 3。產出一律 DRAFT,需人覆核。

執行

python3 skills/accept/tool.py --in <$RUN/verdict.json> --out <$RUN/accepted.json>

獨立重用

skills/accept/ + sopkit/ 可單獨抽出;只要上游給 verdict@1 即可運作。

04 同仓库更多 Skills

s0912758806p
s0912758806p/agentic-sop-to-work

second-opinion

Use when asked to get a second opinion on a finished DRAFT, fact-check or double-check a report's claims against its evidence, sanity-check before sign-off, or red-team an output for fabrication / overreach / invented IDs — even without the word "honesty". Works on an agentic-sop-kit run dir (FULL, uses the run's trace) or any plain document (DEGRADED, you supply the inputs). | 要對一份已完成的 DRAFT 取得「第二意見」、在核准前查核報告主張是否有輸入佐證、抓臆造/灌水/過度宣稱/捏造編號時使用;吃 agentic-sop-kit 的 run 目錄(FULL,用 trace)或任何純文件(DEGRADED,需自備輸入)。不適用:稽核 agent 架構有沒有拆好/是不是 mega-agent(改用 agentic-workflow-audit);審查原始碼品質/bug/資安(改用 /code-review)。本 skill 只查「這次產出的主張 vs. 證據」誠實度,只產 DRAFT,核准永遠人擁有。

★ 211
s0912758806p
s0912758806p/agentic-sop-to-work

plugin-forge

Use when creating, scaffolding, or linting a Claude Code plugin — generate a grammar-conformant plugin skeleton, or validate a plugin / whole marketplace against house invariants (manifest, frontmatter, stdlib-only, hook protocol, test harness). | 要新建 / scaffold 一個 Claude Code plugin,或 lint 既有 plugin / 整個 marketplace 是否符合房規時使用。

★ 211
s0912758806p
s0912758806p/agentic-sop-to-work

good-plugin

demo good plugin

★ 211
s0912758806p
s0912758806p/agentic-sop-to-work

bad-plugin

暂无简介

★ 211
s0912758806p
s0912758806p/agentic-sop-to-work

alcoa-guard

Use when checking a record / dataset / GMP document for ALCOA+ data integrity — attribution, contemporaneity (backdating), completeness, accuracy (recompute / in-spec), consistency — before a human approves. DEGRADED mode lints any CSV/JSON record against a declared .alcoa.json contract; FULL mode snaps onto an agentic-sop-kit run dir. | 要檢查紀錄/資料/GMP 文件的資料完整性(ALCOA+:可歸屬、同步性/防 backdating、完整、正確、一致)時使用。輸出為 DRAFT,人擁有最終判定;確定性檢查只覆蓋可機械驗證的部分,其餘列為人判斷清單,絕不臆造。

★ 211
s0912758806p
s0912758806p/agentic-sop-to-work

sop-flow

Run the agentic-sop-kit workflow (extract→compute→report, or a declared topology) and report the DRAFT result. Works out-of-the-box (bundled demo) even before a project adopts the kit.

★ 211