Menu
Show the current flame-harness pipeline state — phase, round, scores — read-only.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Phase 7 — analyze the game, decide a rewarded-ad strategy, guide manual AdMob ad-unit creation, and inject google_mobile_ads + ATT/UMP consent code.
Phase 3 — define the Flutter design_tokens.dart spec (palette, typography, spacing), the game's art/visual concept, and the asset/audio sourcing plan.
Phase 6 — skeptical QA. Run the game, watch it, then judge against the contract. Default = functional check; --strict adds quality and edge-case passes.
Phase 5 — build the Flame game in 3 gated sub-phases (core loop → systems+components → UI+content), then self-evaluate against the contract.
Phase 10 — upload store text metadata + categories via fastlane, then pause with exact manual steps for the final iOS review submission and Android production promotion.
Phase 8 — bootstrap signing credentials, generate fastlane config from templates, and build + upload signed IPA (TestFlight) and AAB (internal track).
Basé sur la classification professionnelle SOC
| name | flame-harness-status |
| description | Show the current flame-harness pipeline state — phase, round, scores — read-only. |
| argument-hint | |
| allowed-tools | ["Read","Bash","Glob"] |
Read-only status reporter for the flutter-flame-harness pipeline. Reads state.md,
build-log.md, and pipeline-log.md and prints a summary of the current pipeline state.
All file schemas (state.md, build-log.md, pipeline-log.md) are defined in
docs/harness-protocol.md — refer to that document as the single source of truth.
Do not redefine schemas here.
This skill is strictly read-only. It must never modify state.md or any other harness file.
Locate docs/harness/state.md relative to the current project root (the directory that contains
docs/harness/). If state.md does not exist, print:
flame-harness-status: no active pipeline found (docs/harness/state.md missing).
Run /flame-harness <idea> to start a new pipeline.
Then exit without error.
Read docs/harness/state.md and extract the following fields per the schema in
docs/harness-protocol.md §2:
status — running | paused | completedcurrent_phase — name of the currently executing phasecurrent_round — integer round counternext_role — skill role that runs nextpause_reason — "" | rate_limit | manual_action | errorupdated_at — ISO-8601 timestamp of last writeresume_attempts — integerRead docs/harness/build-log.md per the schema in docs/harness-protocol.md §6. Extract the
latest row to obtain the most recent QA score (PASS | FAIL | PARTIAL) and the round number
it corresponds to. If the file is absent or has no data rows, set latest_score to —.
Read docs/harness/pipeline-log.md per the schema in docs/harness-protocol.md §6. Extract the
last 3 rows to give a recent event summary. If the file is absent or empty, skip.
Print the following formatted report (adapt spacing for readability):
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
flame-harness pipeline status
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Status : <status>
Current phase : <current_phase>
Round : <current_round>
Next role : <next_role>
Latest QA : <latest_score>
Last updated : <updated_at>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
If status is paused, append:
⚠ Paused — reason: <pause_reason>
Resume attempts: <resume_attempts>
Run /flame-harness --resume (or /flame-harness-resume) to continue.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
If status is completed, append:
Pipeline completed successfully.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
If there are recent pipeline-log rows, append:
Recent events:
<time> <event> <phase> <details>
<time> <event> <phase> <details>
<time> <event> <phase> <details>
Write, Edit, or any Bash command that modifies a
file. It reads state.md, build-log.md, and pipeline-log.md only.state.md under any circumstances, including error paths.