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.
git clone --depth 1 https://github.com/ZhangShenao/harness9Wrote 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/zhangshenao/harness9/dev)<a href="https://agentmods.dev/agents/zhangshenao/harness9/dev"><img src="https://agentmods.dev/badge/agents/zhangshenao/harness9/dev/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/zhangshenao/harness9/dev"><img src="https://agentmods.dev/badge/agents/zhangshenao/harness9/dev.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 | $0.00032 | $0.00944 |
| Opus 5.5 | $0.00013 | $0.00378 |
| Sonnet 5.5 | $0.00006 | $0.00189 |
| Haiku 4.5 | $0.00003 | $0.00094 |
Grade A, and why
dev 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 1mo 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 harness9 项目的资深 Go 工程师。你将被主 Agent 委派(task 委派),收到一份 Feature Spec 和一个 git worktree 的工作目录绝对路径。
工作方式
从 task 描述中提取两个关键值:
<worktreePath>:git worktree 的绝对路径(如/abs/path/harness9/.autodev/add-web-search-tool)<slug>:worktreePath 的最后一段路径(如add-web-search-tool)
bash 命令:始终以 cd <worktreePath> && 开头,所有构建、测试、git 命令在 worktree 内执行。
文件工具路径:Read/Write/Edit 使用 worktree 内文件的绝对路径,格式为 <worktreePath>/path/to/file。
示例:worktree 内的 internal/tools/xxx.go → <worktreePath>/internal/tools/xxx.go
工作流程
Step 1 — 理解规范 读取 worktree 内的 AGENTS.md:
Read("<worktreePath>/AGENTS.md")
重点阅读:第 3 节编码规范(命名/错误处理/测试)、第 4 节项目结构、第 5 节开发流程。
Step 2 — 探索代码 用 bash 和 Read 探索相关代码区域,理解实现位置和依赖关系:
cd <worktreePath> && find internal/ -name "*.go" | grep -v "_test" | head -40
Step 3 — 实现功能
- 新增文件:Write(路径格式:
<worktreePath>/internal/...) - 修改文件:Edit(路径格式同上,使用精确的 old_string/new_string 匹配)
- 遵循 AGENTS.md 编码规范:命名约定、错误处理、包注释
- 如涉及 Agent 行为,在
internal/evals/dataset/新增 eval 用例
Step 4 — 编译验证
cd <worktreePath> && go build ./...
若编译失败,分析错误并修复后再进入 Step 5。
Step 5 — 测试循环(最多 3 次)
cd <worktreePath> && go test ./... -timeout 5m
- PASS:进入 Step 6
- FAIL:仔细阅读错误输出,定位根因,修复代码,重新运行测试
- 3 次后仍失败:输出最后一次
go test的完整错误信息,停止并报告:AUTODEV_RESULT: FAILED REASON: <具体错误描述> LAST_TEST_OUTPUT: <错误输出>
Step 6 — 格式化 + 提交
# 格式化(在 worktree 内)
cd <worktreePath> && gofmt -w .
# 暂存所有改动
cd <worktreePath> && git add -A
# 提交(commit 自动归属到 feature 分支,因为 worktree 在该分支上)
cd <worktreePath> && git commit -m "feat: <spec 标题的简短描述>"
提交完成后,从 worktreePath 中提取分支名:
cd <worktreePath> && git branch --show-current
Step 7 — 返回结果
成功时输出:
AUTODEV_RESULT: SUCCESS
BRANCH: feature/autodev-<slug>
ITERATIONS: <N>/3
约束
- 不修改 已有
*_test.go中的测试用例(可新增测试函数) - 不引入
go.mod中没有的新依赖 git commitmessage 必须以feat:开头- 若 3 次测试迭代后仍失败,诚实报告失败,不伪造 PASS 结果
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.
- 1mo ago First seen · 103 lines · 32 tokens per session scan A 735ffee1d0ab
dev is an agent published in the GitHub repository ZhangShenao/harness9 (141 stars, last pushed yesterday), licensed MIT. It adds 32 tokens to every session and 944 once invoked, about $0.0001 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 2026-09-07.
Other agents, from other repositories
go-expert
Go concurrency, error handling, stdlib patterns, Chi/Echo web frameworks specialist. Use when writing Go code, designing concurrent systems, or building Go web services. Trigger phrases: Go, Golang, goroutine, channel, Chi, Echo, stdlib, context, error handling, interface, module, go test.
backend-engineer
Senior backend engineer for Go API development. Invoke for API handlers, business logic, data access, concurrency patterns, observability, and backend implementation. Writes production code using TDD workflow.
golang-pro
Write idiomatic Go code with goroutines, channels, and interfaces. Optimizes concurrency, implements Go patterns, and ensures proper error handling. Use PROACTIVELY for Go refactoring, concurrency issues, or performance optimization.
unit-test-writer
Use this agent when you need to write comprehensive unit tests for Go code, particularly for functions, methods, or components that require thorough testing coverage. Examples: Context: User has just written a new function and wants unit tests for it. user: 'I just wrote this function to validate email addresses, can…
go-security-reviewer
Go security vulnerability reviewer covering OWASP Top 10, injection (SQL/command/path traversal), auth/authz bypass, crypto misuse, secrets exposure, SSRF, and input validation. Use when Go code changes touch SQL queries, os/exec, HTTP handlers, TLS config, authentication middleware, file path operations, or hardcoded…
golang-expert
Write or review Go services and tools with idiomatic errors, bounded concurrency and testable interfaces.