survey by supercent-io/skills-template
npx skills add https://github.com/supercent-io/skills-template --skill survey关键词:
survey| 平台:OMC / OMX / OHMG / Claude Code / Codex CLI / Gemini CLI / OpenCode在规划、编码或确定方向之前,先审视全局情况。
/plan、jeo、ralph 或实施之前将研究保存在平台中立的目录中,以便 Claude / Codex / Gemini 都能复用:
.survey/{slug}/
├── triage.md
├── context.md
├── solutions.md
└── platform-map.md # 可选;对于智能体/工具/平台主题是必需的
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
triage.md:问题 / 受众 / 为何是现在context.md:工作流、受影响的用户、变通方案、相关问题、用户声音solutions.md:解决方案清单、类别、频率、缺口、矛盾点、关键洞察platform-map.md:将平台特定的发现标准化为 settings、rules、hooks当主题涉及智能体工具、模型编排、钩子、权限或供应商 API 时,将发现标准化为三个层次:
settings:模型、安全性、温度、最大令牌数、系统提示、供应商参数rules:允许/拒绝/修改策略、护栏、操作约束、审批逻辑hooks:前/后/错误回调、通知处理程序、生命周期自动化、事件触发器在比较 Claude、Codex、Gemini、OMC、OMX 或 OHMG 时使用此抽象。如果供应商的功能可以清晰地映射到这三个层次之一,则不要将其描述为无关的一次性概念。
推荐内部配置文件:
survey_profile:
settings:
search_language: English
output_language: user-language
max_searches_per_lane: 8-10
save_root: .survey
rules:
- facts_only
- no_code_generation
- ask_max_2_triage_questions
- dedupe_solution_names
- include_source_links
hooks:
pre_research:
- normalize_topic
- choose_platform_adapter
post_lane:
- merge_notes
- record_sources
post_run:
- write_summary
- suggest_next_skill
选择最接近的可用编排界面,但保持输出格式一致。
| 平台 | 首选执行形态 | 备注 |
|---|---|---|
| OMC / Claude Code | 4 个并行研究智能体。优先使用面向研究的专业智能体(如果可用);否则使用具有网络搜索功能的通用智能体。 | 将产物保存在 .survey/ 中,而非 .omc/,以便其他平台可以复用。 |
| OMX / Codex CLI | 使用 $research、探索者/规划者风格智能体,或具有网络访问权限的并行工作器。 | 在相关时,将供应商配置转换为 platform-map.md 中的 settings/rules/hooks。 |
| OHMG / Gemini / Antigravity | 使用工作流指南 + 具备 PM/研究能力的智能体或等效的并行通道。 | 保持相同的 4 通道结构和共享产物名称。 |
| OpenCode / 后备方案 | 使用并行的通用网络能力智能体。 | 不要因为缺少专门的智能体名称而阻塞。 |
如果平台特定的专家不存在,则回退到具有网络功能的通用智能体并继续执行。
what、who 或 why 不明确时提问。将请求解析为:
what:需要调研的痛点、想法或能力who:感受到痛点或使用工作流的人why:为何现在很重要在继续之前,检查 .survey/{slug}/triage.md 是否已存在。
保存:
# Triage
- Problem: {what}
- Audience: {who}
- Why now: {why}
同时启动所有通道。
研究:
返回章节:
## Workflow Context## Affected Users## Current Workarounds## Adjacent Problems## User Voices研究:
返回章节:
## Solutions## Frequency Ranking## Categories## Curated Sources研究:
返回章节:
## What People Actually Use## Common Workarounds## Pain Points With Current Solutions## Sources默认模式:
对于智能体 / 模型 / 编排主题,此通道必须将空间标准化为:
## Settings## Rules## Hooks## Platform Gaps在适用时,为 Claude / Codex / Gemini 使用比较表格。
编写 context.md:
# Context: {project name}
## Workflow Context
{when and where the problem shows up}
## Affected Users
| Role | Responsibility | Skill Level |
|------|----------------|-------------|
{rows}
## Current Workarounds
{numbered list with limitations}
## Adjacent Problems
{lettered or bulleted list}
## User Voices
{quotes with source links}
编写 solutions.md:
# Solution Landscape: {project name}
## Solution List
| Name | Approach | Strengths | Weaknesses | Notes |
|------|----------|-----------|------------|-------|
{deduplicated list}
## Categories
{grouped by category}
## What People Actually Use
{behavior summary}
## Frequency Ranking
{most mentioned solutions}
## Key Gaps
{structural gaps not covered today}
## Contradictions
{marketed claims vs user reality}
## Key Insight
{1 paragraph}
当主题涉及平台或智能体时,编写 platform-map.md:
# Platform Map: {project name}
## Settings
| Concern | Claude | Codex | Gemini | Common Layer |
|---------|--------|-------|--------|--------------|
{rows}
## Rules
| Concern | Claude / OMC | Codex / OMX | Gemini / OHMG | Common Layer |
|---------|---------------|-------------|---------------|--------------|
{rows}
## Hooks
| Lifecycle | Claude | Codex | Gemini | Common Layer |
|-----------|--------|-------|--------|--------------|
{rows}
## Platform Gaps
{where abstractions do not align cleanly}
返回简短摘要:
## Survey complete: {slug}
### Context
- {1-2 sentence summary}
- Main workaround: {most common workaround}
### Solution Landscape
- {N} solutions across {M} categories
- Key insight: {one sentence}
- Key gap: {one sentence}
### Files
- `.survey/{slug}/triage.md`
- `.survey/{slug}/context.md`
- `.survey/{slug}/solutions.md`
- `.survey/{slug}/platform-map.md` # if created
除非用户要求,否则不要进入规划或实施阶段。
| 操作 | 说明 |
|---|---|
| 开始调研 | survey <topic> |
| 复用现有结果 | 首先检查 .survey/{slug}/ |
| 搜索策略 | 用英语搜索,用用户的语言撰写 |
| 并行通道 | 背景 + 解决方案 + 行为 + 替代方案/平台映射 |
| 调研后的下一步 | /plan、jeo、ralph,或用户要求时进行实施 |
what / who / why 清晰context.md 已保存solutions.md 已保存platform-map.md 已保存每周安装次数
406
代码库
GitHub 星标数
88
首次出现
2026年3月10日
安全审计
安装于
gemini-cli353
codex347
opencode334
github-copilot329
cursor329
kimi-cli328
Keyword:
survey| Platforms: OMC / OMX / OHMG / Claude Code / Codex CLI / Gemini CLI / OpenCodeSurvey the landscape before planning, coding, or committing to a direction.
/plan, jeo, ralph, or implementationSave research in a platform-neutral directory so Claude / Codex / Gemini can all reuse it:
.survey/{slug}/
├── triage.md
├── context.md
├── solutions.md
└── platform-map.md # optional; required for agent/tooling/platform topics
triage.md: problem / audience / why nowcontext.md: workflows, affected users, workarounds, adjacent problems, user voicessolutions.md: solution inventory, categories, frequency, gaps, contradictions, key insightplatform-map.md: normalize platform-specific findings into settings, rules, hooksWhen the topic involves agent tooling, model orchestration, hooks, permissions, or vendor APIs, normalize findings into three layers:
settings: model, safety, temperature, max tokens, system prompt, provider parametersrules: allow/deny/modify policy, guardrails, action constraints, approval logichooks: pre/post/error callbacks, notify handlers, lifecycle automation, event triggersUse this abstraction when comparing Claude, Codex, Gemini, OMC, OMX, or OHMG. Do not describe vendor features as unrelated one-off concepts if they map cleanly into one of these three layers.
Recommended internal profile:
survey_profile:
settings:
search_language: English
output_language: user-language
max_searches_per_lane: 8-10
save_root: .survey
rules:
- facts_only
- no_code_generation
- ask_max_2_triage_questions
- dedupe_solution_names
- include_source_links
hooks:
pre_research:
- normalize_topic
- choose_platform_adapter
post_lane:
- merge_notes
- record_sources
post_run:
- write_summary
- suggest_next_skill
Select the closest available orchestration surface, but keep the output format identical.
| Platform | Preferred execution shape | Notes |
|---|---|---|
| OMC / Claude Code | 4 parallel research agents. Prefer research-oriented specialist agents when available; otherwise use general-purpose agents with web search. | Keep artifacts in .survey/, not .omc/, so other platforms can reuse them. |
| OMX / Codex CLI | Use $research, explorer/planner-style agents, or parallel workers with web access. | Translate vendor config into settings/rules/hooks in platform-map.md when relevant. |
| OHMG / Gemini / Antigravity | Use Workflow Guide + PM/research-capable agents or equivalent parallel lanes. |
If a platform-specific specialist does not exist, fall back to a general-purpose web-enabled agent and keep going.
what, who, or why is unclear.Parse the request into:
what: the pain point, idea, or capability to surveywho: who feels the pain or uses the workflowwhy: why it matters nowBefore proceeding, check whether .survey/{slug}/triage.md already exists.
Save:
# Triage
- Problem: {what}
- Audience: {who}
- Why now: {why}
Launch all lanes together.
Research:
Return sections:
## Workflow Context## Affected Users## Current Workarounds## Adjacent Problems## User VoicesResearch:
Return sections:
## Solutions## Frequency Ranking## Categories## Curated SourcesResearch:
Return sections:
## What People Actually Use## Common Workarounds## Pain Points With Current Solutions## SourcesDefault mode:
For agent / model / orchestration topics, this lane must instead normalize the space into:
## Settings## Rules## Hooks## Platform GapsUse a comparison table for Claude / Codex / Gemini when applicable.
Write context.md:
# Context: {project name}
## Workflow Context
{when and where the problem shows up}
## Affected Users
| Role | Responsibility | Skill Level |
|------|----------------|-------------|
{rows}
## Current Workarounds
{numbered list with limitations}
## Adjacent Problems
{lettered or bulleted list}
## User Voices
{quotes with source links}
Write solutions.md:
# Solution Landscape: {project name}
## Solution List
| Name | Approach | Strengths | Weaknesses | Notes |
|------|----------|-----------|------------|-------|
{deduplicated list}
## Categories
{grouped by category}
## What People Actually Use
{behavior summary}
## Frequency Ranking
{most mentioned solutions}
## Key Gaps
{structural gaps not covered today}
## Contradictions
{marketed claims vs user reality}
## Key Insight
{1 paragraph}
When the topic is platform or agent related, write platform-map.md:
# Platform Map: {project name}
## Settings
| Concern | Claude | Codex | Gemini | Common Layer |
|---------|--------|-------|--------|--------------|
{rows}
## Rules
| Concern | Claude / OMC | Codex / OMX | Gemini / OHMG | Common Layer |
|---------|---------------|-------------|---------------|--------------|
{rows}
## Hooks
| Lifecycle | Claude | Codex | Gemini | Common Layer |
|-----------|--------|-------|--------|--------------|
{rows}
## Platform Gaps
{where abstractions do not align cleanly}
Return a short summary:
## Survey complete: {slug}
### Context
- {1-2 sentence summary}
- Main workaround: {most common workaround}
### Solution Landscape
- {N} solutions across {M} categories
- Key insight: {one sentence}
- Key gap: {one sentence}
### Files
- `.survey/{slug}/triage.md`
- `.survey/{slug}/context.md`
- `.survey/{slug}/solutions.md`
- `.survey/{slug}/platform-map.md` # if created
Do not move into planning or implementation unless the user asks.
| Action | Instruction |
|---|---|
| Start a survey | survey <topic> |
| Reuse existing results | Check .survey/{slug}/ first |
| Search strategy | Search in English, write in the user's language |
| Parallel lanes | Context + Solutions + Behavior + Alternatives/Platform Map |
| Next step after survey | /plan, jeo, ralph, or implementation if the user asks |
what / who / why are clearcontext.md savedsolutions.md savedplatform-map.md saved for agent/platform topicsWeekly Installs
406
Repository
GitHub Stars
88
First Seen
Mar 10, 2026
Security Audits
Gen Agent Trust HubPassSocketPassSnykWarn
Installed on
gemini-cli353
codex347
opencode334
github-copilot329
cursor329
kimi-cli328
AI 代码实施计划编写技能 | 自动化开发任务分解与 TDD 流程规划工具
41,400 周安装
| Preserve the same 4-lane structure and shared artifact names. |
| OpenCode / fallback | Use generic web-capable agents in parallel. | Do not block on missing specialized agent names. |