본문으로 건너뛰기
cccskills
무료GitHub에 공개

strategize

Prioritize what to work on next, optionally with deep review.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

indigoai-us/hq-core스타 90(GitHub 90 + cccskills 0)

실행 점수 미검증

설치 방법 보기

포함된 파일(2)

  • SKILL.md7.9 KB
  • agents/openai.yaml119 B

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

/strategize - Strategic Prioritization

Answer "what should I work on next?" Quick mode gives a prioritized 3-5 list. Deep mode writes a full strategic review.

Input: $ARGUMENTS

Pipeline: /idea → /brainstorm → /prd → /strategize → /run-project

Step 0: Parse Input & Company Anchor

Check if first word of $ARGUMENTS matches a company slug in companies/manifest.yaml.

How to check: Read companies/manifest.yaml. Extract top-level keys (company slugs). If the first word of $ARGUMENTS exactly matches one:

  1. Set {co} = matched slug. Strip from $ARGUMENTS
  2. Announce: "Anchored on {co}"
  3. Bind the session — run hq core hq-session set company_slug "{co}" now, before any read under companies/{co}/. The scope guard blocks company paths until the session is bound to {co}.

If no match → company resolved in Step 1.

Flag detection: Check remaining args for --deep. If found, set mode = deep, strip flag. Default: mode = quick.

Step 1: Resolve Company

If {co} already set from Step 0: skip to Step 2.

Otherwise, once a company is resolved below, bind the session before reading board.json: hq core hq-session set company_slug "{co}".

Priority order:

  1. cwd inside companies/{slug}/ → infer from path
  2. cwd inside repos/{pub|priv}/{name} → manifest lookup for owning company
  3. Ask via AskUserQuestion: "Which company?" — list slugs that have board.json

Read companies/{co}/board.json. If file doesn't exist: "No board.json for {co}. Run /idea to start one." — exit. If schema_version < 2 or missing: warn "Board is v1 — objective analysis unavailable. Showing project-only view." — proceed with projects only.

Step 2: Load Board Data

From companies/{co}/board.json extract:

  • objectives[] with their key_results[]
  • projects[] — exclude status archived

PRD reads: For each non-archived project where prd_path is set AND status is in_progress or prd_created:

  • Read the prd.json file at the path
  • Extract: story_total = userStories.length, story_done = count where passes === true, prd_status = metadata.status
  • Cap at 15 prd.json reads — prioritize in_progress first, then prd_created
  • If file doesn't exist at prd_path, skip silently (note "PRD not found" in output)

Step 3: Compute Signals

A. STALLED

Projects where:

  • Board status = in_progress
  • Story completion < 25% (or 0 done out of 3+ total)
  • Has prd_path (distinguishes real projects from ideas marked in_progress)

B. READY

Projects where:

  • Board status = prd_created OR prd_status = ready
  • Has valid prd_path
  • Story completion = 0% (not started)

C. ALIGNED-IDEA

Projects where:

  • Board status = idea or exploring
  • Has objective_id (aligned to an objective)
  • No prd_path set

D. COVERAGE-GAP

For each objective: collect all linked project IDs from key_results[].project_ids + projects with matching objective_id. If NO linked project has status in_progress or prd_created → objective has no active work.

E. Capacity (context only)

Count projects with board status in_progress.

Step 4: Score & Rank

Internal scoring (not shown to user):

  • READY: base 80 + 10 if objective_id set
  • ALIGNED-IDEA: base 60 + 20 if brainstorm_path exists
  • STALLED: base 40 + 20 if >50% stories done (almost there), -20 if 0% done and >5 stories
  • COVERAGE-GAP: base 30

Pick top 3-5 across all signal types. Deduplicate (same project can't appear twice). Tiebreak: READY > ALIGNED-IDEA > STALLED > COVERAGE-GAP.

Step 5: Output — Quick Mode

Print to terminal:

STRATEGY — {company} [{date}]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

In flight: {N} active  |  {M} ready  |  {K} stalled

TOP PRIORITIES
━━━━━━━━━━━━━━

1. [READY] {project title}
   Why: {1-sentence reasoning — alignment, effort, momentum}
   State: {status} | {done}/{total} stories | Obj: {obj-title or "unaligned"}
   Next: /run-project {prd-slug}

2. [STALLED] {project title}
   Why: {reasoning}
   State: {status} | {done}/{total} stories | Obj: {obj-title or "unaligned"}
   Next: /run-project {prd-slug}

... (up to 5)

Signal labels: [READY] [STALLED] [IDEA] [GAP]

If COVERAGE-GAP detected but not in top 5: Note: "{obj-title}" has no active projects.

If >10 in-flight projects: Note: {N} concurrent projects — consider completing before starting new work.

Next action routing table:

Project stateSuggested action
idea, no brainstorm/brainstorm {co} {slug}
exploring, has brainstorm/prd {co} {slug}
prd_created or ready/run-project {prd-slug}
in_progress, stalled/run-project {prd-slug} (resume)
COVERAGE-GAP (no project)/idea {co} --objective {obj-id}

Then AskUserQuestion with options:

  1. Act on #1 — {title} ({next-action-verb})
  2. Act on #2 — {title} ({next-action-verb})
  3. Act on #3 — {title} ({next-action-verb})
  4. Run deep review (--deep)

When user picks an option: execute the routed command (e.g., invoke /run-project, /brainstorm, /prd, or /idea).

Step 6: Output — Deep Mode

Everything in Step 5 PLUS generate a markdown report.

Additional analysis:

OKR Progress — For each objective:

  • List KRs with progress (current/target, %). Derived KRs: recompute current from linked project completion counts
  • Status: >=70% → on_track, 40-69% → at_risk, <40% → off_track
  • Flag objectives with ALL KRs off_track or NO KRs as AT-RISK

Project Funnel:

idea: N  →  exploring: N  →  prd_created: N  →  in_progress: N  →  done: N

Assess: top-heavy (many ideas, few executing), balanced, or bottom-heavy.

Unaligned Work — All in_progress or prd_created projects with no objective_id.

Capacity — Flag if in_progress > 5: "High WIP — consider completing before starting."

Write report to: workspace/reports/{co}-strategy-review.md

# {Company} Strategic Review
_Generated {date}_

## Summary
{2-3 sentence executive summary}

## Capacity
{N} projects in flight. {Assessment: focused/overloaded/underleveraged}

## OKR Coverage
| Objective | KRs | Progress | Status |
|---|---|---|---|
| {title} | {count} | {avg %}  | {status} |

### At-Risk Objectives
{List with reasoning, or "None"}

## Project Funnel
{Text funnel + assessment}

## Top Priorities
{Same 3-5 list from quick mode with fuller reasoning}

## Unaligned Work
| ID | Title | Status |
|---|---|---|
{Projects with no objective_id}

## Recommended Actions
{Numbered list matching priorities with full commands}

After writing: qmd update 2>/dev/null || true

Print: Report written: workspace/reports/{co}-strategy-review.md

Then present same AskUserQuestion as quick mode.

Rules

  • Read-only — no board.json writes, no Linear API calls, no PRD modifications
  • 15 prd.json reads max — prevent timeout on large boards
  • No HTML — this command is text/markdown only
  • No brainstorm.md — that is /brainstorm. This command ROUTES to it
  • 1 AskUserQuestion max for clarification (Step 1). 1 for action routing (Step 5/6). Never batched together
  • Skip archived projects — don't surface, don't count
  • Staleness = story completion % — primary signal. updated_at is unreliable (mass-set dates). Say "0/8 stories complete" not "inactive for 30 days"
  • Graceful on missing data — no prd_path → skip story count, show "no PRD". No objectives → skip OKR sections. No KRs → flag as "needs KRs"
  • Company isolation — scope all reads to companies/{co}/. Never mix company data
  • Deep mode writes exactly 1 file — workspace/reports/{co}-strategy-review.md
  • Do NOT use TodoWrite or EnterPlanMode

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

accept

무료

Accept HQ membership invites from a magic link or raw token.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

indigoai-us/hq-core902026년 10월 26일 업데이트

adr

무료

Capture qualifying architecture decisions as ADRs in the right HQ or repo location.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

indigoai-us/hq-core902026년 10월 26일 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.

indigoai-us/hq-core902026년 10월 26일 업데이트

architect

무료

Find shallow modules and propose deepening refactors, ranked with file refs; never edits code. Use when code is hard to change or planning a refactor.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

indigoai-us/hq-core902026년 10월 26일 업데이트

Generate dark ASCII banner graphics for HQ and social launch assets.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

indigoai-us/hq-core902026년 10월 26일 업데이트

Compare approaches to best-in-class prior art, merge existing work, verify cruft removal, and recommend the most durable design before PRD work.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

indigoai-us/hq-core902026년 10월 26일 업데이트

indigoai-us의 스킬 모두 보기

이 스킬의 문제 신고