Przejdź do treści
Biblioteka skilli
Zarządzanie i produkt · Planowanie

flow

/flow

Prowadzi proces buildflow z bramkami od pomysłu do dowodu ukończenia: research, zakres, PRD, ADR, karty, build, przegląd, wdrożenie i retrospektywa.

Repozytorium
flow-skill ↗
Licencja
MIT
Ostatnia zmiana repozytorium
26.10.2026

Ocena i bezpieczeństwo

100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
134
Skrypty
16
Komentarze HTML
0
Adresy URL w SKILL.md
1
Przestarzałość

Brak sygnałów. Repozytorium zmienione 26.10.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co sprawdzić przed instalacją
  • 16 skryptów w paczce. Przejrzyj je przed instalacją, bo Claude może je uruchomić.

Skrypty: adapter-hooks/context_handler.py, eval/fixtures/converge/conv-01/src/app.py, eval/fixtures/converge/gap-01/src/app.py, harness/_db.py, harness/_domain.py, harness/_presence.py, harness/flow_harness.py, harness/graph_executor.py

Co robi i kiedy

Prowadzi proces buildflow z bramkami od pomysłu do dowodu ukończenia: research, zakres, PRD, ADR, karty, build, przegląd, wdrożenie i retrospektywa.

Argumenty

/flow [ resume | next | card | check C-NNN | project-type web|cli|library|skill | mode teach|work | skip <stage> | ready [--js
Oryginalny opis (EN)

Run the buildflow gated build process from idea to real done-evidence. Walk gated stages (Idea->Research->Scope->PRD->ADR->Contract->Cards->Build->Review->Deploy/Ship->Verify->Retro), each with a honest gate that must pass before advancing. Adapts to project type (web|cli|library|skill). Use when starting or driving a real product build, when the user types /flow, /flow next, /flow card, /flow check, or asks to scope/plan/ship a project through gates. Kill at any gate is a valid outcome.

Jak działa

  1. Two-layer harness
  2. PTCONLY
  3. STOP
  4. Run flow.sh first
  5. Three laws
  6. Load table
  7. Dispatch rules
  8. Seams (one-line pointers)
  9. Forbidden
Sekcje z nagłówków SKILL.md, w oryginale · 149 linii, 1 035 słów
Podgląd SKILL.md
---
name: flow
description: Run the buildflow gated build process from idea to real done-evidence. Walk gated stages (Idea->Research->Scope->PRD->ADR->Contract->Cards->Build->Review->Deploy/Ship->Verify->Retro), each with a honest gate that must pass before advancing. Adapts to project type (web|cli|library|skill). Use when starting or driving a real product build, when the user types /flow, /flow next, /flow card, /flow check, or asks to scope/plan/ship a project through gates. Kill at any gate is a valid outcome.
user-invocable: true
when_to_use: "User wants to build a real product end-to-end with discipline (idea -> a deployed URL for web, or installs+runs for a CLI/library/skill), or types any /flow command, or asks for a gated build process, scope decision, contract-first plan, or card-based shipping."
argument-hint: "[ resume | next | card | check C-NNN | project-type web|cli|library|skill | mode teach|work | skip <stage> | ready [--json] | workspace add|list|enter|remove|check|doctor | auto [--parallel] [stop] | attest semantic|live-verify|status|recover | doctor | retro | eval | clarify | converge | or just say what you want in plain language ]"
keywords: [flow, buildflow, gate, build, ship, scope, prd, contract, card, deploy, vertical-slice, cli, library, skill, worktree, parallel-agents, workspace, multi-agent]
license: MIT
metadata:
  author: flow-skill
  version: "1.0.0"
  attribution: "Methodology from ai20k-build-phase/buildflow (Tony, arealisticdreamer.com); durable-layer ancestry and improve-ritual spirit from repository-harness (protocol v1 EOL — flow-owned fork); agent patterns from claudekit-engineer, BMAD-METHOD. v0.22 concierge from BMAD bmad-help; forge-idea from bmad-forge-idea (MIT, BMad Code LLC)."
---

# /flow — buildflow gated build harness

Idea to **world-state done-evidence** (deployed URL for web; install+run for cli/library/skill), not paperwork. Each stage has an artifact and a **GATE**.
**Kill at any gate is a valid, honored outcome.**

```
Idea -> Research -> Scope -> PRD -> ADR -> Contract -> Cards -> Build -> Review -> Deploy -> Verify-live -> Retro
|------------------ planning (files in flow/) ------------------|  |------- shipping (inside cards/) -------|
```

## Two-layer harness

`/flow` is two layers. Both must agree; the script can pass hollow content.

1. **Mechanical — `runner/flow.sh`** (deterministic, exit 0/1). Stage/card lifecycle plus
   cheatable checks: unchecked boxes (including leftover `- [ ]` under `## Open decisions`),
   `[FILL]`, card status, empty done-evidence, plus content scanners on 01/02/03/05
   (typed-web entry floor, L-above-A, PRD adjectives, OD cap). Always run it first — its
   exit code is ground truth, never host judgment. Open-decision leftover boxes are the
   same scanner; `clarify` only prints them — not a second gate.
2. **Semantic — this skill.** The script cannot tell a real competitor quote from a
   fabricated one, or a grade-laundered C feature from an honest B. After mechanical
   PASS, apply `references/gate-examples.md` before letting the operator advance.

## PTC_ONLY

This file is always-on. Load **one** extra reference for the verb in play (load table)
plus that verb's row in `references/command-dispatch.md`. Do not load the rest until
needed. Typed verbs always win over chat.

## STOP

- Runner reports **BLOCKED by another session's lock** → STOP. Never `FLOW_FORCE` a live
  session; concurrent runs corrupt the plan. Set `FLOW_SESSION_ID` per session.
- Mechanical **FAIL** → STOP. Relay exact line numbers. In `teach`, never tick a box or
  write an artifact on the operator's behalf.
- Security-class skip/debt → **HALT** until the operator accepts the exposure in `DEBT.md`.
- Never mark a card `done` without pasted world-state evidence.
- Never edit `_templates/` or `runner/flow.sh` during a project run.
- `NEXT_VERB=` / JSON `next_verb` is advisory; never auto-exec `next`, `card`,
  `card-start`, `check`, `fix-gate`, `auto`, or `skip`.

## Run `flow.sh` first

From the **project root** (where `flow/` and `cards/` live):

```
# macOS / Linux / Windows Git Bash:
bash <skill-dir>/runner/flow.sh <command>

# Windows PowerShell or cmd (INCLUDING inside Codex): use the .cmd launcher, NOT bare bash.
<skill-dir>\runner\flow.cmd <command>
```

Bare `bash` on Windows/Codex is usually WSL and cannot read `C:/` paths. Always
`runner/flow.cmd` from a Windows shell; it locates Git Bash. Call `flow.sh` directly
only when `bash` is Git Bash.

**Flow v1 runtime requirement (pre-release target):** product operations require Python
3.7 or newer (standard library only), resolved as `python3` then `python`. Missing,
below-floor, or nonworking interpreters refuse progression; a nonempty
`FLOW_HARNESS_DISABLE` also refuses product operations. Help and `doctor` remain available
for diagnosis. The graph executor owns ordering; the mechanical runner remains the gate
verdict authority. This contract is not shipped until implementation, migration/rollback,
review and CI acceptance are complete; see the [v1 migration guide](https://github.com/manhquydev/flow-skill/blob/master/docs/migration-v1.md).

`<skill-dir>` is the install home (`~/.claude/skills/flow`, `~/.codex/skills/flow`,
`~/.agents/skills/flow`, Antigravity `~/.gemini/antigravity-cli/skills/flow` (CLI) /
`~/.gemini/config/skills/flow` (IDE), or project `.claude/skills/flow`). Override with
`FLOW_PROJECT_ROOT`. Antigravity (`agy` CLI / IDE): same `SKILL.md` bundle; run
`agy inspect` to confirm it loaded.

**One session per project.** `flow/.lock` auto-reclaims after `FLOW_LOCK_TTL` (default
900s). Mutating commands refuse a fresh foreign lock. Export a stable `FLOW_SESSION_ID`.
`FLOW_FORCE=1` only for a lock known dead; `unlock` clears it.

## Three laws

1. **Inspect first.** Competitors, live systems, code. Evidence, not vibes.
2. **Contract is the seam.** Stage 05 before any code. Backend builds TO it, UI consumes
   FROM it. Amend the contract, then code. Honor a shape now (null/stub) even when the
   value ships later.
3. **Done = proof in the world.** Name type-aware done-evidence up front: deployed URL
   for web; install+run for cli/library/skill. "Tests pass" / "code merged" are
   mid-pipeline, never done.

## Load table

| In play | Load |
|---|---|
| typed verb | `references/command-dispatch.md` (that row only) |
| plain language | `references/concierge.md`, `references/flow-catalog.tsv` |
| next PASS / check semantic | `references/gate-examples.md` |
| card / build session | `law/CLAUDE.md` |
| UI | `law/DESIGN.md` |
| auto | `references/auto-run.md` |
| attest / receipts | `references/attestations.md` |
| mode work | `references/mode-work.md` |
| eval | `references/gate-eval.md` |
| Idea/Scope ritual | `references/forge-idea.md` |
| engines | `references/codex-integration.md`, `references/antigravity-integration.md`, `references/claudekit-skills.md` |
| parallel occupancy | `references/host-agnostic-parallel.md` |
| agents | `references/agent-detection.md`, `references/agent-stage-mapping.md` |
| skip / halt | `references/debt-and-halts.md` |
| host context adapters (opt-in) | `references/host-hooks.md` |
| harness | `harness/README.md` |
| retro | `law/RETRO.md` |

## Dispatch rules

1. **Entering a project mid-cycle?** Fresh session and existing `flow/` or `cards/` → run
   resume first (read-only session-story: last session, in-flight + dwell, gate state,
   one `NEXT ->` line). Skip only with live context this session.
2. **Always call `flow.sh` first** and relay exit + output. Lock BLOCKED → STOP.
3. **On next:** FAIL → stop, offer help, never author/tick in teach. PASS → semantic
   challenge in `gate-examples.md` (the one extra file). Do not silently advance hollow
   content; do not silently block a sound artifact.
4. **On card/check:** `law/CLAUDE.md` — one card per session, only `## Allowed files`,
   contract is the seam, done = world-state proof. Run recall first; apply its output.
5. **Mode `work`:** interview once, draft 00–05, pause only for scope sign-off, same
   gates as teach.
6. **Never** edit `_templates/` or `runner/flow.sh` during a project run. Read any file
   the runner just created before editing it.

## Seams (one-line pointers)

- Verb map: `references/command-dispatch.md` (no Commands table in this file).
- Chat front door: `references/concierge.md` + `references/flow-catalog.tsv`.
- Idea/Scope persona ritual: `references/forge-idea.md` (adapted from BMAD-METHOD,
  `bmad-forge-idea`, MIT, opt-in, never a gate condition).
- Codex second engine: `references/codex-integration.md`.
- Antigravity Gemini-3 **third engine**: `references/antigravity-integration.md`
  (confirm load with `agy inspect`).
- ck/ak skill layer: `references/claudekit-skills.md`.
- Parallel occupancy: `references/host-agnostic-parallel.md`.
- Receipts: `references/attestations.md`.
- `harness/` is **flow-owned**. Improve via **R-IMPROVE-HARNESS** in
  `references/native-rituals.md`. Never `story update --status implemented` — use
  `story complete --proof-source`.

## Forbidden

- Checking a gate box or writing a planning artifact on the operator's behalf (teach).
- Setting a card `done` without pasted world-state evidence.
- Building two cards in one session, or in parallel before `ready` marks them safe.
- Frontend code before the UI mock card is approved.
- Editing `_templates/` or `runner/flow.sh` during a project run.
Pliki w paczce (134)
  • SKILL.md 9 KB
  • _templates/00-idea.md 1 KB
  • _templates/00-inspect.md 3 KB
  • _templates/01-research.md 3 KB
  • _templates/02-scope.md 5 KB
  • _templates/03-prd.md 4 KB
  • _templates/04-adr.md 1 KB
  • _templates/05-contract.md 6 KB
  • _templates/card.md 1 KB
  • _templates/constitution.md 1 KB
  • adapter-hooks/claude-settings-snippet.json 1 KB
  • adapter-hooks/context_handler.py 41 KB
  • adapter-hooks/flow-context-handler 1 KB
  • eval/manifest.tsv 1 KB
  • eval/fixtures/b00_supported/flow/00-inspect.md 1 KB
  • eval/fixtures/b00_unchecked/flow/00-inspect.md 1 KB
  • eval/fixtures/converge/manifest.tsv 1 KB
  • eval/fixtures/converge/conv-01/cards/C-001.md 1 KB
  • eval/fixtures/converge/conv-01/flow/03-prd.md 1 KB
  • eval/fixtures/converge/conv-01/src/app.py 1 KB
  • eval/fixtures/converge/gap-01/cards/C-001.md 1 KB
  • eval/fixtures/converge/gap-01/cards/C-002.md 1 KB
  • eval/fixtures/converge/gap-01/flow/03-prd.md 1 KB
  • eval/fixtures/converge/gap-01/flow/05-contract.md 1 KB
  • eval/fixtures/converge/gap-01/src/app.py 1 KB
  • eval/fixtures/f01a/flow/01-research.md 4 KB
  • eval/fixtures/f01b/flow/01-research.md 4 KB
  • eval/fixtures/f02a/flow/02-scope.md 3 KB
  • eval/fixtures/f02b/flow/02-scope.md 4 KB
  • eval/fixtures/f05a/flow/05-contract.md 1 KB
  • eval/fixtures/f05b/flow/05-contract.md 1 KB
  • eval/fixtures/f06_injected/flow/05-contract.md 1 KB
  • eval/fixtures/f06_sound/flow/05-contract.md 1 KB
  • eval/fixtures/f07_acyclic/flow/05-contract.md 1 KB
  • eval/fixtures/f07_missing_ledger/flow/05-contract.md 1 KB
  • eval/fixtures/fcda/.flow/.gitkeep 1 KB
  • eval/fixtures/fcda/cards/C-001.md 2 KB
  • eval/fixtures/fcdb/.flow/.gitkeep 1 KB
  • eval/fixtures/fcdb/cards/C-001.md 1 KB
  • eval/fixtures/fcdc/.flow/.gitkeep 1 KB
  • eval/fixtures/fcdc/cards/C-001.md 1 KB
  • eval/fixtures/fcdd/cards/C-001.md 1 KB
  • eval/fixtures/fcde/cards/C-001.md 2 KB
  • eval/fixtures/routing/manifest.tsv 1 KB
  • eval/fixtures/routing/states/state-any-midproject.txt 1 KB
  • eval/fixtures/routing/states/state-cards-todo.txt 1 KB
  • eval/fixtures/routing/states/state-empty.txt 1 KB
  • eval/fixtures/routing/states/state-idea-stage.txt 1 KB
  • eval/fixtures/routing/states/state-review-passed.txt 1 KB
  • eval/fixtures/routing/states/state-review-pending.txt 1 KB
  • eval/fixtures/routing/states/state-scope-stage.txt 1 KB
  • eval/replay/README.md 3 KB
  • harness/GAP-MATRIX-0.1.17.md 3 KB
  • harness/README.md 14 KB
  • harness/_db.py 21 KB
  • harness/_domain.py 7 KB
  • harness/_presence.py 4 KB
  • harness/flow_harness.py 85 KB
  • harness/graph_executor.py 59 KB
  • harness/graph_ids.py 1 KB
  • harness/graph_predicates.py 2 KB
  • harness/repo_map.py 4 KB
  • harness/pins/flow-topology.sha256 1 KB
  • harness/pins/harness-cli-v0.1.17.sha256sums 1 KB
  • harness/schema/001-init.sql 6 KB
  • harness/schema/002-story-verify.sql 1 KB
  • harness/schema/003-tool-registry.sql 1 KB
  • harness/schema/004-intervention.sql 1 KB
  • harness/schema/005-tool-extensions.sql 1 KB
  • harness/schema/009-accessed-count.sql 1 KB
  • harness/schema/010-usage-event.sql 2 KB
  • harness/schema/011-usage-gate-reason.sql 1 KB
  • harness/schema/012-usage-ephemeral.sql 1 KB
  • harness/schema/014-graph-executor.sql 3 KB
  • law/CLAUDE.md 9 KB
  • law/CODING.md 2 KB
  • law/DESIGN.md 7 KB
  • law/RETRO.md 1 KB
  • playbooks/README.md 2 KB
  • playbooks/cloudflare-ai-gateway-qwen.md 17 KB
  • playbooks/docker-deploy-stale-cache.md 2 KB
  • playbooks/rag-lite-postgres-cloudflare.md 9 KB
  • playbooks/windows-gitbash-jsonl-telemetry.md 3 KB
  • references/adversarial-review.md 13 KB
  • references/agent-detection.md 7 KB
  • references/agent-stage-mapping.md 10 KB
  • references/antigravity-integration.md 6 KB
  • references/artifact-lifecycle.md 3 KB
  • references/attestations.md 11 KB
  • references/auto-run.md 11 KB
  • references/clarify.md 9 KB
  • references/claudekit-skills.md 12 KB
  • references/codex-integration.md 7 KB
  • references/command-dispatch.md 10 KB
  • references/concierge.md 5 KB
  • references/converge.md 5 KB
  • references/debt-and-halts.md 3 KB
  • references/design-review-checklist.md 3 KB
  • references/flow-catalog.tsv 3 KB
  • references/flow-topology.json 4 KB
  • references/forge-idea.md 5 KB
  • references/gate-00.md 2 KB
  • references/gate-01.md 1 KB
  • references/gate-02.md 1 KB
  • references/gate-03.md 1 KB
  • references/gate-04.md 1 KB
  • references/gate-05.md 4 KB
  • references/gate-assess.md 1 KB
  • references/gate-card.md 2 KB
  • references/gate-consistency.md 3 KB
  • references/gate-constitution.md 1 KB
  • references/gate-debt.md 1 KB
  • references/gate-eval.md 21 KB
  • references/gate-examples.md 5 KB
  • references/gate-rules.md 2 KB
  • references/gate-shared.md 2 KB
  • references/ground-truth-gates.md 6 KB
  • references/host-agnostic-parallel.md 8 KB
  • references/host-hooks.md 6 KB
  • references/loop-harness-2026-principles.md 3 KB
  • references/mode-work.md 3 KB
  • references/native-rituals.md 9 KB
  • references/pin-v1.txt 1 KB
  • references/project-types.md 3 KB
  • references/stage-state-machine.md 4 KB
  • references/test-impact.md 6 KB
  • references/ui-patterns-tcr.md 3 KB
  • references/workspace-rules.md 6 KB
  • runner/attestations.sh 60 KB
  • runner/doctor-checks.tsv 1 KB
  • runner/doctor_helper.py 20 KB
  • runner/flow.cmd 1 KB
  • runner/flow.sh 334 KB
  • runner/installer_helper.py 31 KB

Wersje

RepozytoriumWynikLicencja
Ta strona · manhquydev/flow-skill100MIT
lifecycle-innovations-limited/claude-ops100MIT
simota/agent-skills100MIT
justinedevs/meta-architect100MIT
cofin/flow100Apache-2.0
panayiotism/claude-harness100MIT
Panza-Labs/floorboss96Apache-2.0
ohdearquant/lionagi95Apache-2.0
AO-Cyber-Systems/devflow-claude92MIT

Migawka bazy z 13.11.2026 · wersja paczki c76e6117d4b1 (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 100gate
    Przeprowadza bramkę cyklu życia projektu, zadając pytania techniczne i biznesowe, zbiera dowody i wydaje PASS, HOLD lub KILL.
  • 90prd-os
    Prowadzi formalny, etapowy proces: zamienia szkic w PRD, sprawdza go przeglądem, dzieli na zgłoszenia i realizuje je z kontrolą zakresu.
  • 90prd-os
    Prowadzi formalny, etapowy proces: zamienia szkic w PRD, sprawdza go przeglądem, dzieli na zgłoszenia i realizuje je z kontrolą zakresu.
  • 100ship
    Prowadzi pomysł albo istniejący projekt do gotowości produkcyjnej: specyfikacja, budowa, przegląd UX, QA i raport.
  • 85research-to-ship
    Prowadzi od researchu sprawdzonych praktyk do planu, ograniczonego przeglądu i projektu, z bramkami zatwierdzenia po każdym etapie.
  • 90research-to-ship
    Prowadzi etapową pracę od badania dobrych praktyk, przez plan i przegląd, do projektu wykonywanego zgodnie z dev-flow.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.