본문으로 건너뛰기
cccskills
무료GitHub에 공개

research-note

Synthesis step in the multi-angle technical research pipeline. Receives structured outputs from all four research angles (api-state, ecosystem-research, impact-measurement, codebase-auditor), applies cross-angle signal weighting and conflict resolution, and produces a single synthesized Research section. Invoked by the technical-researcher orchestrator after all angle skills complete. Returns content to the orchestrator — does not write to the backlog.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(1)

  • SKILL.md4.5 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Research Note

Synthesize multi-angle research outputs into a single Research section with signal weighting and conflict resolution. Return content to the calling orchestrator — the orchestrator writes to the backlog.

Step 1 — Receive Angle Outputs

Accept the structured report text from each angle that ran:

  • api-state — API surface, stability, deprecations
  • ecosystem-research — compatibility, dependencies, community health
  • impact-measurement — quantitative benchmarks, adoption, performance data
  • codebase-auditor — behavioral contracts, coding conventions, workflow insertion points, agent data flow maps

Identify and isolate the Gaps section from each report — you will need it in Step 4.

Step 2 — Cross-Angle Signal Detection

For each factual claim that appears in 2 or more independent angle reports, mark it [CROSS-ANGLE].

A claim qualifies when:

  • Two angles independently reached the same conclusion from different sources
  • Neither angle cites the other's output as its source

Correct: mark [CROSS-ANGLE] only for genuinely independent corroboration. One angle quoting another angle's cited source is not independent corroboration — mark it only in the section of the angle that sourced it independently.

Step 3 — Conflict Detection and Resolution

For each claim where two angles contradict each other, mark it [CONFLICT] and apply the resolution procedure:

  1. One angle has a primary source citation, the other does not → the cited claim wins. State the resolution explicitly.
  2. Both angles have citations and still contradict → mark [UNRESOLVED] and surface it.

Every contradiction must reach a stated resolution or an [UNRESOLVED] marker. Surface all conflicts in the Conflicts section of the output.

Step 4 — Gap Aggregation

Collect the Gaps section from every angle report. Separate into two priority tiers:

  • Cross-angle gaps — appear in 2 or more angle reports. Highest priority for follow-up.
  • Single-angle gaps — appear in only one angle report.

Write "None identified." when a tier has no entries. Both tiers must appear in the output.

Step 5 — All-INCONCLUSIVE Gate

Before synthesizing, check: did all angles produce only Gaps with no substantive findings?

flowchart TD
    Q{Did at least one angle<br>produce substantive findings?}
    Q -->|Yes| Synth[Proceed to Step 6 — synthesize]
    Q -->|No — all angles returned<br>INCONCLUSIVE| Blocked["Return BLOCKED response (see format below)<br>Do not produce a synthesized note"]

BLOCKED response format:

BLOCKED — all research angles returned INCONCLUSIVE. No verifiable findings were produced.
Gaps: {aggregated gap list}
Escalate to human for manual research or alternative sources.

Step 6 — Synthesize the Research Note

Produce the final Research section in this exact format. Apply [CROSS-ANGLE] and [CONFLICT] markers inline within each angle's section where the marked claims appear. Synthesize only what the angles produced — add no new claims.

### Research — {technology/feature} — {date}

Research method: {N}-angle independent fan-out. [CROSS-ANGLE] marks findings confirmed by 2+ independent angles. [CONFLICT] marks contradictions between sources.

### API State
{content from api-state angle, with [CROSS-ANGLE] markers applied}

### Ecosystem and Compatibility
{content from ecosystem-research angle, with [CROSS-ANGLE] markers applied}

### Quantitative Impact
{content from impact-measurement angle, with [CROSS-ANGLE] markers applied}

### Codebase Audit
{content from codebase-auditor angle, with [CROSS-ANGLE] markers applied — omit section if codebase-auditor did not run}

### Cross-Angle Signals
| Finding | Angles | Confidence |
|---|---|---|
| {claim} | {angle-X + angle-Y} | High |

### Conflicts
[CONFLICT] {claim A — source} vs {claim B — source} — [RESOLVED: reason | UNRESOLVED]

### Gaps (cross-angle)
{Gaps that appeared in 2+ angle reports — highest priority for follow-up}

### Gaps (single-angle)
{Gaps from individual angle reports}

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Add automated documentation updater to any Claude skill. Creates a Python sync script that downloads upstream docs, processes markdown for AI consumption, and maintains local cache with configurable refresh. Collects template variables, then delegates implementation through 5-phase workflow. Use when adding auto-updating reference documentation to plugins or skills.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Jamie-BitFlight/claude_skills672026년 10월 20일 업데이트

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Jamie-BitFlight/claude_skills672026년 10월 20일 업데이트

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Jamie-BitFlight/claude_skills672026년 10월 20일 업데이트

Browser automation for AI agents using the agent-browser CLI and Playwright. Use when navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, logging into sites, or automating any browser task. Triggers on "open a website", "fill out a form", "click a button", "scrape data", "test this web app", "automate browser actions", or any programmatic web interaction request.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Jamie-BitFlight/claude_skills672026년 10월 20일 업데이트

Runs the description-drift experiment — spawns all Claude Code agents simultaneously to collect self-reported capabilities, then compares them against static frontmatter descriptions to reveal how reliable orchestrator routing based on descriptions actually is. Use when measuring description drift across the agent fleet, re-running the capability collection experiment, analyzing a specific agent's self-reported capabilities, or auditing whether frontmatter descriptions accurately reflect agent behavior.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Jamie-BitFlight/claude_skills672026년 10월 20일 업데이트

Create or adapt Claude Code agent definitions. Use when creating an agent, changing subagent configuration, selecting agent scope, or designing a specialized delegation role.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Jamie-BitFlight/claude_skills672026년 10월 20일 업데이트

Jamie-BitFlight의 스킬 모두 보기

이 스킬의 문제 신고