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 unknowbug/anchorlaw --skill anchor.judgegit clone --depth 1 https://github.com/unknowbug/anchorlawWrote 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/unknowbug/anchorlaw/anchor.judge)<a href="https://agentmods.dev/skills/unknowbug/anchorlaw/anchor.judge"><img src="https://agentmods.dev/badge/skills/unknowbug/anchorlaw/anchor.judge/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/unknowbug/anchorlaw/anchor.judge"><img src="https://agentmods.dev/badge/skills/unknowbug/anchorlaw/anchor.judge.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 | $0.00046 | $0.01980 |
| Opus 5 | $0.00023 | $0.00990 |
| Sonnet 5 | $0.00009 | $0.00396 |
| Haiku 4.5 | $0.00005 | $0.00198 |
Grade A, and why
anchor.judge 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- anchor-judge — 95% identical, 5 lines differ
How it starts
The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
anchor.judge — 审查角色（subprocess）
Protocol: spec/protocol-v0.20.md §15.4 (Consistency Contract), §16 (Host Integration) Layer: 执行角色（非 §14 动作 skill，不占 manifest 名额） Execution: subprocess（隔离）
角色契约（§15.4 审查门）
- 本角色在隔离子进程中运行：工作上下文绝不进入主会话。
- 只出审查意见，绝不直接修改产物 status。 状态提升由主会话/宿主人类裁决。
confirmed只能由宿主侧人类授予——审查意见只能建议candidate。
双模式（v0.13 §15.1/§15.4）
v0.10 起 judge 角色有两种用法，分别对应流水线的驱动与把关：
- 驱动者（主会话化身，inline 判据） — 主会话以 Judge 角色全程驱动四段流水线（v0.11）： 输入契约接受（外部需求 + 规范 → 推导验收判据）→ 规范审查 → 计划批准 → 模块并入审查 → 交付总审。持有验收判据，对照判据裁决（正常路径不依赖轮次计数；同一判据 3 轮未达标 → 整个流程硬停止，提交 Review 情况与问题详细报告交人类判定，§15.4 C）。
- 隔离验收者（subprocess，独立视角） — 在两个关键节点 MUST 派隔离 judge subprocess 独立验收： 模块并入前（stage 3）+ 交付前（stage 4）。自评≠审查：主会话 Judge 的点头是驱动， 隔离 judge 的裁定是门。
强制触发点（v0.9；v0.10 扩展）
judge 不是只在任务收尾跑——以下决策点 MUST 触发审查（工作流计划在规划阶段就 MUST 预置这些 judge 步骤）：
confirmed授予前 — 人类授予confirmed前 MUST 已有本角色的审查意见；无审查意见时candidate是最高可达状态。- 重大转向前 — 结案重开（推翻已确认结论）/ 根因定论（如「无 bug」断言）/ 范围决策（增减验证范围），行动前 MUST 先审查被推翻/被断言的结论。
- 阶段结论授予
candidate— SHOULD 触发审查；未审查即授予时，缺失审查意见 MUST 在产物中标注。
（v0.11 输入契约 + 四段流水线触发点，见 协议 §15.4）： 4. 输入契约接受 — Judge 接受外部需求文档 + 技术约束规范输入并推导验收判据（stage 0，无输入不得开工）； 5. 规范审查 — Judge 审过实施规范才进规划（stage 1 终）； 6. 计划批准 — Judge 批准模块划分（stage 2 终）； 7. 模块并入 — Judge 审每模块并入/打回（stage 3）； 8. 交付验收 — Judge 总审 + 隔离 judge 独立验收后授权交付（stage 4）。
自评 ≠ 审查：主 Agent/执行者的自检与验证是证据，不是本角色的审查——审查 MUST 由独立执行者完成（本角色即独立 subprocess）。
计划预置（§15.4 Plan-time placement）：工作流计划（todo 列表）MUST 在规划阶段预置本角色的触发步骤；到达触发点而无预置审查步骤时，MUST 停下补加审查再继续，不得事后补跑当作合规。
意见分级（v0.9 §15.4 Verification termination gates）
审查意见分恰好三层——只有第 1 层阻塞提交，其余层不阻塞（记录去向即可）：
- blocking — 仅限三类机械判据：① 测试失败；② 编译失败；③ 协议声称与实现相矛盾。除此之外任何意见不得归入 blocking。
- should-fix — 建议性修改：记录处置去向（采纳 / 记为 debt），不阻塞提交。
- info — 记录/观察：可忽略。与本地可验证证据冲突的意见（如编码误报）归入 info 并记录排除理由；排除误报的验证不得超过 1 轮。
对应义务（对主 Agent，§15.4）：任何采纳意见导致的改动 → MUST 重新跑外部测试集（批准计划 + 仓库既有测试集，不含验证阶段自加测试），测试绿即验证完成；v0.12 起同一验收判据迭代 3 轮未达标 → 不得继续打回——v0.15 起 MUST 整个流程硬停止：不得继续修复/重新分类/自行 re-review，必须提交 Review 情况与问题详细报告交人类判定（判据错 → §12 挑战或人类修正；方向错 → 回规划；或人类另行裁决），不依赖自觉判断「是否持续」；意见与证据冲突 → 走 §12 或记录排除，不进验证循环重检。
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 Changed 57005127b379
- 10d ago First seen · 85 lines · 46 tokens per session scan A bf58faed9e22
anchor.judge is a skill published in the GitHub repository unknowbug/anchorlaw (5 stars, last pushed 2d ago), licensed MIT. It adds 46 tokens to every session and 1,980 once invoked, about $0.0002 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
solid-principles
SOLID principles checklist with Java examples. Use when a class has too many responsibilities, an abstraction leaks, or a dependency points the wrong way, and when the user asks about Single Responsibility, Open/Closed, Liskov, Interface Segregation or Dependency Inversion. For naming, duplication and method length…
security-audit
Java security checklist covering OWASP Top 10, input validation, injection prevention, and secure coding. Works with Spring, Quarkus, Jakarta EE, and plain Java. Use when reviewing code security, before releases, or when user asks about vulnerabilities.
issue-triage
Triage and categorize GitHub issues with priority labels. Use when user says "triage issues", "check issues", "review open issues", or during regular maintenance of GitHub issue backlog.
ring:migrating-to-lib-observability
Migrating a Lerian Go app off lib-commons observability imports (deprecated shims or removed APIs) to lib-observability via a fixed mapping table, then bumps go.mod and validates the build; ring:backend-go applies the edits. Covers log/zap/runtime/assert, opentelemetry/tracing, HTTP middleware, context helpers, and…
ring:searching-code
Forensic code search and analysis with optional Chain of Draft (CoD) ultra-concise mode. Five-phase methodology (clarification, planning, execution, analysis, synthesis) with severity assessment. Use for targeted investigation of specific patterns, bugs, or vulnerabilities. Skip for broad architecture mapping (use…
ring:exploring-codebases
Exploring a codebase across phases: scopes the target, detects architecture, components, and layers, deep-dives each discovered perspective, then synthesizes findings into actionable guidance with file:line evidence. Use to understand how a feature or system works before planning changes, or to orient on an unfamiliar…