ccoalm/ccl-skills / skills/multi-perspective-research
multi-perspective-research
Verified调研一个主题 / 深度调研 / deep research / 调研某个产品·公司·赛道 / 多视角研究 / 帮我研究一下 X / 写作前调研 / 领域·组织研究(不喂决策)/ research a topic → run a structured research loop(视角枚举 → 矛盾图 → 合成简报 → 自评)with evidence-grounding discipline, instead of one-question surface answers; another installed deep-research skill may serve as a search tool inside this loop, not replace it. Skip:一查便知的单点事实问题 → 直接回答不套流程;请求点名了候选让选、或用裁决措辞要结论(选哪个 / 要不要上 X / 该不该做 / 可不可行)→ 先 `product-rd-workflow`(它可回调本技能产调研底稿);纯主题/领域/产品/写作准备调研留在本技能;拷问已有方案 → `grill-me`;bug 根因 → `defect-diagnosis`;复盘沉淀 → `skill-extraction-workflow`。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 32.6 KB |
| agents/openai.yaml | config | 394 B |
| references/public-data-acquisition.md | markdown | 46.4 KB |
| references/public-disclosure-channels.md | markdown | 13.2 KB |
| references/research-prompts.md | markdown | 4.3 KB |
| scripts/AGENTS.md | markdown | 1.7 KB |
| scripts/test-public-data-acquisition-recipes.shexecutable | script | 18.5 KB |
scripts/AGENTS.md
View on GitHub# skills/multi-perspective-research/scripts Agent Contract
This directory contains deterministic regression checks for executable recipes
embedded in the multi-perspective research references. The repository-wide
contract remains authoritative: [root `AGENTS.md`](../../../AGENTS.md). This
file only narrows that contract; it does not relax its safety, authority,
owner-routing, worktree-isolation, or validation gates.
Rules:
- Execute snippets extracted from the owning reference instead of maintaining a
second implementation that can drift from the published recipe.
- Keep the inline `MANIFEST` in
`test-public-data-acquisition-recipes.sh` exhaustive: every runnable code
block must be marked as executed or explicitly exempt with a concrete
environment or integration reason, and an unregistered block must fail the
check.
- Use synthetic fixtures and local stubs. Do not require live network access,
production data, credentials, or private endpoints.
- Preserve fail-closed assertions for acquisition, parsing, completeness, and
cleanup behavior. A failed check must return non-zero.
- Keep shell checks portable across supported macOS and Linux environments.
Use trap-based cleanup on normal, failed, and catchable interrupted exits.
Validation, run from the repository root:
- `bash skills/multi-perspective-research/scripts/test-public-data-acquisition-recipes.sh`
- `shellcheck -S warning skills/multi-perspective-research/scripts/test-public-data-acquisition-recipes.sh`
- `bash skills/product-rd-workflow/scripts/check-agent-contract-coverage.sh --repo . --enforce`
If `shellcheck` is unavailable, report that validation as inconclusive rather
than passing.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |