论文生成
- 作者仓库星标 39
- 作者更新于 实时读取
- 作者仓库 awesome-omni-skill
- 领域
- 工程开发
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @diegosouzapw · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 需简单配置
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- 未声明（默认跨平台）
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- Shell 执行
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生，可能与作者实际意图存在差异。
需要注意： 未限定 allowed-tools，默认拥有全部工具权限。
---
name: trace
description: Trace design decisions and concepts through session history, handoffs, and git. Triggers: "trace…
category: 工程开发
runtime: 无特殊运行时
---
# trace 输出预览
## PART A: 任务判断
- 适用问题：代码实现、重构、调试或代码审查。
- 输入要求：目标材料、限制条件、期望输出和验收方式。
- 证据边界：围绕“When to Use / Execution Steps / Step 1: Classify Target Type”读取原文规则，不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断：确认你的需求是否属于代码实现、重构、调试或代码审查，并标出输入、限制和预期结果。
- **02** 执行计划：优先按“When to Use / Execution Steps / Step 1: Classify Target Type”拆成步骤，说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果：给出可复制的命令、文件改动、检查清单或内容草稿，并说明如何继续迭代。
- **04** 风险边界：结合 读取文件、写入/修改文件、执行终端命令、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件、执行终端命令；主要在本地完成；通常不需要额外 API Key。
- 先小样例验证，再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。原文出现了 `/provenance`、`/trace` 这类斜杠命令；如果你的 Agent 支持命令触发，优先用命令开场，再补充目标和边界。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是：读取文件、写入/修改文件、执行终端命令。
先用一个小任务确认它会围绕“When to Use / Execution Steps / Step 1: Classify Target Type”工作；涉及文件或命令时，先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作；如果输出泛泛而谈，就补充输入、边界和验收标准后重跑。
---
name: trace
description: Trace design decisions and concepts through session history, handoffs, and git. Triggers: "trace…
category: 工程开发
source: diegosouzapw/awesome-omni-skill
---
# trace
## 什么时候使用
- 把工程方向的常用动作沉淀成 Agent 可调用的技能 适合处理工程开发场景下的代码实现、调试、重构、测试或代码审查，核心价值是把输入、判断、执行、验证和交付边界固定下来，避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代…
- 面向代码实现、重构、调试或代码审查，优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「When to Use / Execution Steps / Step 1: Classify Target Type」组织步骤，不把推断写成作者事实。
- 读取文件、写入/修改文件、执行终端命令；主要在本地完成；通常不需要额外 API Key。
- 先跑小样例，确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown，不编造命令、平台或依赖。作者原文负责流程事实；仓库文件负责来源和命令；流狐只补充适用场景、限制和质量判断。
skill "trace" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> When to Use / Execution Steps / Step 1: Classify Target Type
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件、执行终端命令 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
}Trace Skill
Quick Ref: Trace design decisions through CASS sessions, handoffs, git, and artifacts. Output:
.agents/research/YYYY-MM-DD-trace-*.md
YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it.
When to Use
- Trace HOW architectural decisions evolved
- Find WHEN a concept was introduced
- Understand WHY something was designed a certain way
- Build provenance chain for design decisions
For knowledge artifact lineage (learnings, patterns, tiers), use /provenance instead.
CLI dependencies: cass (session search). If cass is unavailable, skip transcript search and rely on git log, handoff docs, and .agents/ artifacts for decision tracing.
Execution Steps
Given /trace <concept>:
Step 1: Classify Target Type
Determine what kind of provenance to trace:
IF target is a file path (contains "/" or "."):
→ Use /provenance (artifact lineage)
IF target is a git ref (sha, branch, tag):
→ Use git-based tracing (Step 2b)
ELSE (keyword/concept):
→ Use design decision tracing (Step 2a)Step 2a: Design Decision Tracing (Concepts)
Launch 4 parallel search agents (CASS, Handoff, Git, Research) and wait for all to complete.
Backend: Agents use Task(subagent_type="Explore") which maps to task(subagent_type="explore") in OpenCode. See skills/shared/SKILL.md ("Runtime-Native Spawn Backend Selection") for the shared contract.
Read references/discovery-patterns.md for agent definitions and prompts.
Step 2b: Git-Based Tracing (Commits/Refs)
Read references/discovery-patterns.md for git-based tracing commands.
Step 3: Build Timeline
Merge results from all sources into a single chronological timeline (oldest first). Deduplicate same-day/same-session events. Every claim needs a source citation.
Step 4: Extract Key Decisions
For each event in timeline, identify:
- What changed: The decision or evolution
- Why: Reasoning if available
- Who: Session/author/commit author
- Evidence: Link to source (session path, file, commit)
Step 5: Write Trace Report
Write to: .agents/research/YYYY-MM-DD-trace-<concept-slug>.md
Read references/report-template.md for the full report format and deduplication rules.
Step 6: Report to User
Tell the user:
- Concept traced successfully
- Timeline of evolution (key dates)
- Most significant decisions
- Location of trace report
- Related concepts to explore
Handling Edge Cases
Read references/edge-cases.md for handling: no CASS results, no handoffs, ambiguous concepts (>20 results), and all-sources-empty scenarios. General principle: continue with remaining sources and note gaps in the report.
Key Rules
- Search ALL sources - CASS, handoffs, git, research
- Build timeline - chronological evolution is the goal
- Cite evidence - every claim needs a source
- Handle gaps gracefully - not all concepts are in all sources
- Write report - trace must produce
.agents/research/artifact
Relationship to /provenance
| Skill | Purpose | Input | Output |
|---|---|---|---|
/provenance |
Artifact lineage | File path | Tier/promotion history |
/trace |
Design decisions | Concept/keyword | Timeline of evolution |
Use /provenance for: "Where did this learning come from?"
Use /trace for: "How did we decide on this architecture?"
Examples
# Trace a design decision
/trace "three-level architecture"
# Trace a role/concept
/trace "Chiron"
# Trace a pattern
/trace "brownian ratchet"
# Trace a feature
/trace "parallel wave execution"Tracing an Architectural Decision
User says: /trace "agent team protocol"
What happens:
- Agent classifies target as concept (not file path or git ref)
- Agent launches 4 parallel agents: CASS search, handoff search, git log search, research artifact search
- CASS finds 8 sessions mentioning "agent team", handoff finds 2 docs, git finds 3 commits, research finds 1 analysis
- Agent builds chronological timeline from 2026-01-15 (first mention) to 2026-02-08 (latest update)
- Agent extracts 5 key decisions: initial SendMessage design, TeamCreate addition, deliberation protocol, in-process mode, delegate mode
- Agent writes trace report to
.agents/research/2026-02-13-trace-agent-team-protocol.mdwith full timeline and citations
Result: Complete evolution timeline showing how agent team protocol developed across 7 sessions with source citations.
Tracing from Git Commit
User says: /trace abc1234
What happens:
- Agent detects git ref format (short sha)
- Agent runs git-based tracing commands to get commit details, changed files, related commits
- Agent uses
git log --grepto find related work - Agent searches
.agents/for contemporary research/plans - Agent builds timeline focused on that specific change
- Agent writes report showing commit context, what changed, why (from commit message and related docs)
Result: Trace report links commit to broader design context from surrounding artifacts.
Troubleshooting
| Problem | Cause | Solution |
|---|---|---|
| CASS returns no results | Session search not installed or query too specific | Check which cass. If missing, skip CASS and rely on handoffs/git/research. Try broader query terms. |
| Timeline has gaps | Not all decisions documented in searchable artifacts | Note gaps in report. Suggest interviewing team members or checking Slack/email archives for missing context. |
| Too many results (>50 matches) | Very broad concept or high-frequency term | Read references/edge-cases.md for ambiguous concept handling. Narrow query or filter by date range. Ask user for more specific aspect to trace. |
| Empty trace report (all sources failed) | Concept genuinely undocumented or typo | Verify spelling. Try synonyms. Report to user: "No documented history found. This may be a new concept or may need different search terms." |
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核