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 agentmods add skills/zhangshenao/harness9/debugging-guidenpx skills add ZhangShenao/harness9 --skill debugging-guidegit clone --depth 1 https://github.com/ZhangShenao/harness9What 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 | $0.00022 | $0.00883 |
| Opus 5 | $0.00011 | $0.00441 |
| Sonnet 5 | $0.00004 | $0.00177 |
| Haiku 4.5 | $0.00002 | $0.00088 |
Grade A, and why
debugging-guide 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
harness9 调试指南
诊断顺序
遇到问题时,按以下顺序排查:
1. 编译错误
go build ./...
常见原因:
- 未使用的 import → 删除或添加
_blank import - 类型不匹配 → 检查接口实现是否完整
- 循环依赖 → 将接口定义移到使用者侧包中
2. 测试失败
# 详细输出
go test -v ./internal/engine/
# 单个测试
go test -v -run TestAgentLoop ./internal/engine/
# 带 race detector
go test -race ./...
3. 运行时 panic
查看完整 goroutine stack:
go run ./cmd/harness9 2>&1 | head -100
nil pointer panic 通常来自:
- 未初始化的 map(用
make(map[K]V)初始化) - 接口值为 nil 但调用了方法
4. Agent 行为异常
LLM 不调用工具: 检查工具的 Definition() 描述是否清晰,JSON Schema 是否正确。
工具执行失败: 查看 ToolResult.IsError 和 Output 字段,错误信息会回传给 LLM。
无限循环: 检查 WithMaxTurns 配置,默认 50 Turn。
常用调试技巧
打印 System Prompt
在 internal/context/builder.go 的 Build() 方法末尾临时添加:
fmt.Fprintf(os.Stderr, "=== SYSTEM PROMPT ===\n%s\n===================\n", prompt)
检查工具注册
在 registry.Execute 前打印可用工具列表:
for _, def := range registry.GetAvailableTools() {
fmt.Fprintf(os.Stderr, "tool: %s\n", def.Name)
}
Provider 请求/响应
如需查看实际 API 请求,在 internal/provider/openai.go 中打印消息列表。
harness9 特有问题
Anthropic Provider:user/assistant 必须严格交替
症状:400 Bad Request 或 invalid_request_error
原因:Anthropic Messages API 禁止连续 assistant 消息。
修复:检查 contextHistory 的消息顺序,确保 system→user→assistant→user→assistant 交替。
路径沙箱拒绝访问
症状:工具返回 路径超出工作区范围 或类似错误
原因:路径包含 ../ 或绝对路径指向启动目录之外。
修复:Agent 应使用相对于启动目录的路径,如 internal/engine/agent_loop.go 而非 /absolute/path/...。
Skills 未加载
症状:Agent 不知道有 Skills 可用
检查:
skills/目录是否在项目根目录(启动目录)下- 每个 Skill 是否在独立子目录中,且子目录内有
SKILL.md文件 SKILL.md是否包含name和descriptionfrontmatter 字段- 启动日志中是否有
[skills]warn 输出
正确的目录结构示例:
skills/
├── go-coding-standards/
│ └── SKILL.md ← 必须是这个文件名
└── debugging-guide/
└── SKILL.md
go vet 常见 warning
| Warning | 含义 | 修复 |
|---|---|---|
printf 格式不匹配 |
%s 传入了非 string 类型 |
修正格式或类型 |
unreachable code |
return 后有代码 | 删除死代码 |
loop variable captured |
goroutine 捕获了循环变量 | 传参而非捕获 |
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.
- 2d ago First seen · 122 lines · 22 tokens per session scan A 2875c5183add
debugging-guide is a skill published in the GitHub repository ZhangShenao/harness9 (135 stars, last pushed 3d ago), licensed MIT. It adds 22 tokens to every session and 883 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…