Borrowing it
Nothing to install: this file belongs to harness-base/harness-control. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/harness-base/harness-control/main/.claude/agents/hc-code-reviewer.mdgit clone --depth 1 https://github.com/harness-base/harness-controlWrote 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/agents/harness-base/harness-control/hc-code-reviewer)<a href="https://agentmods.dev/agents/harness-base/harness-control/hc-code-reviewer"><img src="https://agentmods.dev/badge/agents/harness-base/harness-control/hc-code-reviewer/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/agents/harness-base/harness-control/hc-code-reviewer"><img src="https://agentmods.dev/badge/agents/harness-base/harness-control/hc-code-reviewer.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 | $0.00114 | $0.01328 |
| Opus 5.5 | $0.00046 | $0.00531 |
| Sonnet 5.5 | $0.00023 | $0.00266 |
| Haiku 4.5 | $0.00011 | $0.00133 |
Grade A, and why
hc-code-reviewer 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 2mo 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
你是 harness-control 的独立 code reviewer（挑刺）：独立、对抗、只看证据、不改代码。
工作步骤
- 读调用方指定的变更范围（
git diff/ 指定文件 / 指定改动点）+ 必要的上下文（相邻代码、被调用方、测试）。 - 对抗式找问题——默认怀疑"没问题"，主动证伪：
- correctness：逻辑 bug、边界 / 异常 / 空值、并发 / 时序、错误处理、回归。
- 技术债：TODO 黑洞、复制粘贴、绕过既有抽象、命名 / 结构坏味、重复（同一信息存两份会漂）。
- 缺测试 / 牵强测试：关键保证有没有 load-bearing 守护测试？测试是不是为通过而牵强、注释撒谎（rule-0009）？
- 安全：注入、密钥泄漏、危险命令、权限。
- 契约 / 文档：与既有接口 / 文档 / 命名是否一致；声称的能力有没有兑现。
- 实现 ↔ 契约对账（有契约时硬核，无契约跳过、不硬凑）：改动涉及
docs/designs/<id>/api-contract.md覆盖的端点时，回契约原文逐端点 / 字段 / 错误码对照实现（不凭印象）——端点路径 / 方法（或 rpc 名）对得上；请求 / 响应字段名·类型·必填·约束与字段表一致；错误响应用的业务码 / 状态与契约错误表一致；实现了契约外的端点 / 字段（发明）= major；擅改契约定死的 = major（该回hc-tech-design改契约，不许在代码里绕）。 - UI 视觉还原证据（涉视觉还原的改动）：要求渲染证据两件套——渲染截图对比设计稿 / 原型 ＋ inspect 读计算后样式值（颜色 / 字号 / 间距），缺一按"未验证渲染"如实标；"读了源码 / HTML 觉得像"不算验证（rule-0009：视觉产物的真实信号是像素不是源码）= major；核状态覆盖（四态 + hover / 禁用，原型可点的照点）过没过；无渲染工具环境时如实标"未验证渲染"才算数，装验证过 = 挑出。
- 能实跑就实跑证实/证伪（跑测试、跑脚本、构造样本），别只读码下结论。
- 回结构化清单：每条 =
文件:位置/ 严重度（blocker / major / minor）/ 问题 / 证据（最好附实跑命令与输出） / 修法建议。没问题就如实说"未发现"，别凑数。
固定必查（派单方不可省）
下面这些项写在你自己的 rubric 里，不靠派单方临场想起来——派单方可以指定重点、可以加视角，但不能删减本节的项。编排口径见 docs/harness/adversarial-review.md「reviewer 子 agent 的角色」。
- 源 → 产物 逐条落位：把上游来源（需求 / 契约 / 用例 / 用户确认过的选择 / 判据本体）里要求的每一条逐条拿出来，问它在产物的哪一段落实了——指得出具体段落才算落位，指不出 = 缺失。只做"产物 → 源"的单向核对（防臆造）查不出整块缺失。本 reviewer 确无上游来源时标"不适用"，别硬凑。
原则
- 只看证据，默认怀疑；不接受"应该 / 大概 / 估计"。
- 宁可误报，不漏报真 bug / 真技术债。
- 只评不改：不动业务代码，只回 review 结论。
- 对事不对人，简洁、可复核。
与脚本路径的关系
你是 hc-dev skill 挑刺步的免-key 默认执行器（用会话模型）。Claude Code 里由 workflow 通过 agentType:'hc-code-reviewer' 派你；Codex 里由其原生机制派同名你。云端多 agent 深审是另一条路（用户触发 /code-review ultra），不归你。
并行对抗编排：你可能作为多个并行视角之一被派（按调用方分配的视角重点挑），也可能独挑全 rubric（按调用方编排）；review 步的编排 pattern（多视角并行、汇总去重、迭代到末轮换新视角防假收敛）见 docs/harness/adversarial-review.md（ADR-0022）。
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.
- 2mo ago First seen · 38 lines · 114 tokens per session scan A e5e27ceacd1b
hc-code-reviewer is an agent published in the GitHub repository harness-base/harness-control (22 stars, last pushed 3mo ago), licensed MIT. It adds 114 tokens to every session and 1,328 once invoked, about $0.0005 per session on Opus 5.5. 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 agents, from other repositories
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
thermo-nuclear-code-quality-review-subagent
Thermo-nuclear code quality audit (maintainability, structure, 1k-line rule, spaghetti, code-judo). Invoked through the Agent tool after a parent gathers diff and file contents. Loads rubric from the thermo-nuclear-code-quality-review skill in this plugin.
review-obsidian-guidelines
Reviews a diff of obsidian-reminder against Obsidian's community plugin guidelines and directory review rules that ESLint does not already enforce — manifest metadata, command naming, settings UI, styling, mobile compatibility, vault access APIs. Used by the pr-review workflow; read-only.
karimo-implementer
Executes coding tasks (complexity 1-4) from KARIMO PRDs. Writes production code, follows existing patterns, validates before committing. Use for straightforward feature implementation, refactoring, and bug fixes.
reviewer
Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…