[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78BVKT0GPX91BX5B4GY2EWC":3,"skill-same:01M78BVKT0GPX91BX5B4GY2EWC":52},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":14,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M78BVKT0GPX91BX5B4GY2EWC","do-flow",null,"Run the doflow feature chain end to end, from idea to reviewed code. Use when the user wants the whole lifecycle run, says 'build this feature end-to-end', 'run the doflow chain' or 'take this from idea to implementation', or \u002Fdo classified the request as a feature. A plain change request that names no \u002Fdo-* skill goes to \u002Fdo first, which decides whether it needs this flow. Skip it for one phase on its own (that phase's \u002Fdo-* skill), single-file fixes, questions and reviews.","khoavu882\u002Fdo-flow","khoavu882","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F47971376?v=4","01M78BT78KNM1P7A984W2D3JJB","Khoa Vu","01M78BT7957N1KCW7C9CVX4FV1",0,"MIT","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo-flow\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: do-flow\ndescription: \"Run the doflow feature chain end to end, from idea to reviewed code. Use when the user wants the whole lifecycle run, says 'build this feature end-to-end', 'run the doflow chain' or 'take this from idea to implementation', or \u002Fdo classified the request as a feature. A plain change request that names no \u002Fdo-* skill goes to \u002Fdo first, which decides whether it needs this flow. Skip it for one phase on its own (that phase's \u002Fdo-* skill), single-file fixes, questions and reviews.\"\nargument-hint: \"[feature description] [--from brainstorm|design|plan|implement|test|review]\"\neffort: high\n---\n\n# do-flow\n\nRuns a DoFlow workflow end-to-end without a manual `\u002Fdo-X` at every stage. The task's **class**\nselects the workflow; the resolved workflow's stages name which existing skills run, in what order,\nand where the run stops. `do-flow` does not replace or modify those skills — it invokes their own\nBehavioral Flows in sequence. `do-constitution` belongs to no class; invoke it directly when you\nneed to set or amend repo-level rules.\n\n**You propose the class; the runtime validates it.** Never the reverse. A class you assumed, or one\nthe runtime rejected and you replaced with the familiar default, runs a confident workflow for the\nwrong task.\n\n## Invocation\n```text\n\u002Fdo-flow [feature description] [--from brainstorm|design|plan|implement|test|review]\n```\n\n## Behavioral Flow\n\n1. **Resolve state** — run the resolver. Every DoFlow runtime call in this skill goes through the\n   runtime seam. Resolve it **once** here and reuse `$DOFLOW` for every later call in this skill:\n\n```bash\n# Resolve the DoFlow runtime: nearest project install wins, then the global one.\nD=$PWD; while [ \"$D\" != \u002F ] && [ ! -x \"$D\u002F.doflow\u002Fscripts\u002Fdoflow\u002Fbin\u002Fdoflow-run\" ]; do D=$(dirname \"$D\"); done\nDOFLOW=\"$D\u002F.doflow\u002Fscripts\u002Fdoflow\u002Fbin\u002Fdoflow-run\"\n[ -x \"$DOFLOW\" ] || DOFLOW=\"$HOME\u002F.doflow\u002Fscripts\u002Fdoflow\u002Fbin\u002Fdoflow-run\"\n[ -x \"$DOFLOW\" ] || { echo \"doflow: no runtime found in any .doflow\u002F above $PWD, nor at $HOME\u002F.doflow. Run: npx @khoavu882\u002Fdoflow install -t \u003Charness>\" >&2; exit 2; }\n```\nRun every command below from the project root — the walk-up starts at `$PWD`. On exit 2, print the message verbatim and stop; it names every path searched.\n\n```bash\n\"$DOFLOW\" paths --json\n```\n`feature_slug` `null` with a non-empty `candidate_slugs` is an unresolved choice, not \"no active feature\": ask one option per entry, re-resolve with `\"$DOFLOW\" paths --json --slug=\"\u003Cchosen>\"`, and use that slug for the rest of this flow. `\u002Fdo-flow` passing `--slug` already resolves it — no prompt then.\n\n2. **Propose one class** — read the request and name exactly one class id, using the cues and the\n   confusable pairs in this skill's own `references\u002Ftask_classes.md`. State the class and the\n   signal it rests on in one line. If two classes fit equally, do **not** take the longer one: show\n   what each would run —\n   `\"$DOFLOW\" workflow --task-class \"\u003Ccandidate>\" --json`, quote its `stageIds` — and ask via\n   the question tool, one option per candidate.\n\n3. **Validate the class through the runtime** — a proposal is not a selection until this returns:\n\n```bash\n\"$DOFLOW\" classify --task-class \"\u003Cproposed>\" --calling-skill do-flow --json\n```\nBranch on the returned `outcome` field, not the exit code.\n- **`ACCEPTED`** — the returned `workflow` is this run's plan of record; read `stages`, `gates` and `handoff` off it rather than from memory.\n- **`REJECTED`** — **stop.** Print `message` verbatim (it already names `validClasses` and any `suggestions`), ask the user to choose from `validClasses`, then re-validate. Never substitute `feature`.\n  A rejection may be about **you** rather than the class (`reason: caller-not-a-stage`). Then the fix is to propose one of the classes in `fit.hostingClasses`, or to hand the work to the skill this class names for the stage you meant — not to re-propose the same class.\n- **Exit 2** — surface the message verbatim and stop.\n\nEvery stage this run enters comes from the accepted `workflow`'s `stages`; a phase the workflow does\nnot declare is not run here, however familiar it is from the `feature` chain.\n\n4. **Announce the selection** — one line before anything runs:\n   `\u003CtaskClass> → \u003Cworkflow.name>: \u003CstageIds joined by →>`, plus the gate ids if `gates` is\n   non-empty. This is the user's chance to correct a misclassification cheaply.\n\n5. **Pick the starting stage.**\n   - `feature`: resume from the first missing chain artifact — `!has_requirement` → the discovery\n     stage; `has_requirement && !has_design` → design; `has_design && !has_plan` → planning; all\n     three present → go straight to the pre-implementation gate in step 7.\n   - Every other class: start at the first entry in `stages`.\n   - `--from \u003Cphase>` overrides the above to deliberately re-run a stage. Match the phase to the\n     stage whose `skill` is that phase's skill (`brainstorm` → `do-brainstorm`, `design` →\n     `do-design`, `plan` → `do-plan`, `implement` → the stage with `mutatesSource: true`, `test` →\n     `do-test`, `review` → `do-code-review`). No stage matches in this workflow → stop and list the\n     stage ids it does have. Two stages match — `bug` and `refactor` each run `do-test` twice — →\n     ask which one.\n\n6. **Run the stages in order**, invoking each stage's `skill` Behavioral Flow directly.\n   - `optional: true` — decide from that stage's own `purpose`, then say which way you went and\n     why. Never drop an optional stage silently.\n   - `readinessTemplate` is non-null — consult readiness before entering the stage:\n     `\"$DOFLOW\" readiness --task-class \"\u003CtaskClass>\" --task-id \"\u003Ctask id>\" --mode workflow --json`.\n     Both `--task-class` and `--task-id` are required; omitting either exits 2 and names the valid\n     set. Act on the returned `stageEntry.decision` — the entry policy is the runtime's, not this\n     page's: `ENTER` → enter the stage; `GATHER_FIRST` → report the missing items and gather them\n     first, do not enter; `ASK_USER` → ask the owed decision and wait; `STOP` → stop and surface\n     the conflicted claim.\n   - `readinessTemplate` is `null` — enter the stage. Do not consult readiness \"to be safe\": a\n     workflow whose `requiresImplementationReadiness` is `false` has no implementation to be ready\n     for and declares no template by design (`references\u002Ftask_classes.md`). Calling it anyway\n     invents a gate the registry does not declare.\n   - After each stage, report one line: stage id, skill, and the artifact path or result it\n     produced — so the user can follow without intervening.\n\n7. **Stop at the gates the resolved workflow declares** — read them from `workflow.gates`, each\n   attached to a stage by `afterStage`. `trigger: always` stops every time; any other trigger stops\n   only when that named condition actually holds. Ask with the gate's own `prompt` via\n   the question tool. For the `feature` workflow's three, which behave as they always have:\n   - **`gate-0`** (after discovery, trigger `unresolved-clarifications`) — a safety net, not the\n     normal path: `do-brainstorm` resolves ambiguities to zero via its own loop, so this fires only\n     if a `[NEEDS CLARIFICATION]` marker survived an aborted session. Ask one question per marker\n     (grouped if closely related), patch the answers into `requirement.md` directly, and do not\n     re-run `do-brainstorm`. Then follow steps 1-4 of the decision step in the guidance\n     tree's `references\u002FWORKFLOW_HANDOFF.md` (stage `discovery`, channel `question`) so the answers\n     are registered and cited; do-brainstorm already recorded discovery's handoff, so record none.\n   - **`gate-a`** (after planning) — the conversational checkpoint on top of the already-existing\n     advisory hook (`pre-implement-gate.sh`; its scope is in `references\u002FDOFLOW_CHAIN.md`), which\n     still reports a missing artifact. Options: `Proceed` \u002F `Let me review the artifacts first` \u002F `Stop here`.\n   - **`gate-b`** (after review) — no enforcement hook exists for this today; `\u002Fdo-git` is a\n     separate, always-manually-invoked skill. Frame the question by the review's own approval\n     status: on `CHANGES REQUESTED`, foreground addressing the findings rather than offering a bare\n     proceed.\n   A workflow with an empty `gates` array stops at none of these. It still stops on a readiness\n   block, on an ambiguous answer, and on any stage skill's own stopping rule.\n\n   **Record every answer that is a decision.** Asking is not recording: a gate this skill asked but\n   never wrote back stays open in the run, so the stage skill that owns it asks the same question\n   again — the user answers twice for one decision. Immediately after the user answers, write it:\n   ```bash\n   \"$DOFLOW\" orchestrate --action decide-gate --task-id \"\u003Cslug>\" --gate \"\u003Cgate id>\" --decision approve|reject --note \"\u003Cthe user's own answer, one line>\" --json\n   ```\n   Map each gate's own option set onto the two decisions the runtime accepts, and never onto a third:\n   - **`gate-a`** — `Proceed` is `approve`; `Stop here` is `reject`. `Let me review the artifacts\n     first` is **neither**: it is a pause, not a decision, so call nothing and actually pause — hold\n     the run at this gate and re-ask once the user comes back, exactly as step 8's ambiguous-answer\n     rule already requires.\n   - **`gate-b`** — an actual \"proceed to commit\u002Fmerge\" answer is `approve`; an explicit stop is\n     `reject`. Anything that is not a real decision — \"let me address the findings first\" on a\n     `CHANGES REQUESTED` review, most often — is neither, and gets no `decide-gate` call. `reject`\n     terminates the run outright, which is a different answer from \"not now\".\n   - **`gate-0`** — only in the aborted-session branch above, where this skill patched\n     `requirement.md` itself: once every surviving marker is resolved, `approve`. If any marker is\n     still open, leave the gate alone.\n   This call is advisory to the trail, not to the chain: if it fails for a reason outside this flow's\n   control (an unwritable local state directory, say), report the failure plainly and carry the\n   user's answer forward anyway — the answer is what governs whether the next stage runs, and the\n   record of it is what keeps the next stage from asking again.\n\n   When the answer also carries a choice beyond approve or reject and `paths --json` reports\n   `has_decisions: true`, register it too, once `decision --action list`\n   shows no live decision with that statement; the receiving stage's batch then leaves it out:\n   ```bash\n   \"$DOFLOW\" decision --action add --topic \"\u003Ckey>\" --statement \"\u003Ctext>\" --channel gate --stage \"\u003Cstage the gate follows>\" --rationale \"\u003Ctext>\" --json\n   ```\n\n8. **Never skip a gate on an ambiguous answer** — an unanswered or unclear response to the\n   question tool means stop and ask again, per RULE_04 in `rules\u002FRULES.md`. When the\n   terminal stage completes, state `workflow.handoff` as the next step rather than continuing into\n   it.\n\n## Conditional research within a feature\n\nBefore a feature stage relies on an external fact that local or user evidence cannot establish, or when the developer explicitly asks a research question, let that stage follow `..\u002Fdo-document\u002Freferences\u002Fembedded-research.md`. The stage opens a same-task `research-request`, records a source-linked finding or gap, then resumes itself without adding a seventh workflow stage. On resume, read request status and context pack. A blocking unanswered question prevents stage completion; neither research nor a retrieved page approves any existing gate. Feature discovery and design run the ideation pass; otherwise search only for a gap or an explicit request.\n\n## Boundaries\n**Will:**\n- Propose a task class, validate it through the runtime, and run the resolved workflow's stages in\n  order without a manual re-invocation at each boundary.\n- Stop only where the resolved workflow says to: its declared gates, a readiness block, or an\n  ambiguous answer.\n- Record each gate answer it collects through `orchestrate --action decide-gate`, so the stage skill\n  that owns that gate finds it already resolved instead of asking the user the same question twice.\n- Compose with the existing `pre-implement-gate.sh` advisory hook rather than bypass or duplicate it.\n  That hook keys on branch and artifact state, not on class, so it can advise against a `bug` or\n  `trivial-edit` workflow's edits on a branch that already holds an incomplete feature dir —\n  surface its message as written.\n\n**Will Not:**\n- Run any workflow for a class the runtime rejected, or replace a rejected class with `feature`.\n- Infer a class silently when two fit, or read a class the user stated as a suggestion.\n- Consult implementation readiness for a stage that declares no readiness template, or invent a\n  readiness template for a class that has none.\n- Modify any stage skill, or invoke `do-constitution` — it is standalone and in no class.\n- Add a hook-level enforcement gate for commit\u002Fmerge — `gate-b` is conversational only. A\n  hook-level version would be a separate, future proposal.\n- Silently proceed past a gate on an ambiguous or missing answer.\n- Record a pause or a deferral as a gate decision — `Let me review the artifacts first` and \"let me\n  address the findings first\" are neither `approve` nor `reject`, and forcing them into one writes a\n  decision the user did not make.\n\n**Next Step:** whatever the resolved `workflow.handoff` names — for every class that ends in a\nchange, `\u002Fdo-git` to commit\u002Fmerge once the last gate is cleared.\n","# do-flow\n\nRuns a DoFlow workflow end-to-end without a manual `\u002Fdo-X` at every stage. The task's **class**\nselects the workflow; the resolved workflow's stages name which existing skills run, in what order,\nand where the run stops. `do-flow` does not replace or modify those skills — it invokes their own\nBehavioral Flows in sequence. `do-constitution` belongs to no class; invoke it directly when you\nneed to set or amend repo-level rules.\n\n**You propose the class; the runtime validates it.** Never the reverse. A class you assumed, or one\nthe runtime rejected and you replaced with the familiar default, runs a confident workflow for the\nwrong task.\n\n## Invocation\n```text\n\u002Fdo-flow [feature description] [--from brainstorm|design|plan|implement|test|review]\n```\n\n## Behavioral Flow\n\n1. **Resolve state** — run the resolver. Every DoFlow runtime call in this skill goes through the\n   runtime seam. Resolve it **once** here and reuse `$DOFLOW` for every later call in this skill:\n\n```bash\n# Resolve the DoFlow runtime: nearest project install wins, then the global one.\nD=$PWD; while [ \"$D\" != \u002F ] && [ ! -x \"$D\u002F.doflow\u002Fscripts\u002Fdoflow\u002Fbin\u002Fdoflow-run\" ]; do D=$(dirname \"$D\"); done\nDOFLOW=\"$D\u002F.doflow\u002Fscripts\u002Fdoflow\u002Fbin\u002Fdoflow-run\"\n[ -x \"$DOFLOW\" ] || DOFLOW=\"$HOME\u002F.doflow\u002Fscripts\u002Fdoflow\u002Fbin\u002Fdoflow-run\"\n[ -x \"$DOFLOW\" ] || { echo \"doflow: no runtime found in any .doflow\u002F above $PWD, nor at $HOME\u002F.doflow. Run: npx @khoavu882\u002Fdoflow install -t \u003Charness>\" >&2; exit 2; }\n```\nRun every command below from the project root — the walk-up starts at `$PWD`. On exit 2, print the message verbatim and stop; it names every path searched.\n\n```bash\n\"$DOFLOW\" paths --json\n```\n`feature_slug` `null` with a non-empty `candidate_slugs` is an unresolved choice, not \"no active feature\": ask one option per entry, re-resolve with `\"$DOFLOW\" paths --json --slug=\"\u003Cchosen>\"`, and use that slug for the rest of this flow. `\u002Fdo-flow` passing `--slug` already resolves it — no prompt then.\n\n2. **Propose one class** — read the request and name exactly one class id, using the cues and the\n   confusable pairs in this skill's own `references\u002Ftask_classes.md`. State the class and the\n   signal it rests on in one line. If two classes fit equally, do **not** take the longer one: show\n   what each would run —\n   `\"$DOFLOW\" workflow --task-class \"\u003Ccandidate>\" --json`, quote its `stageIds` — and ask via\n   the question tool, one option per candidate.\n\n3. **Validate the class through the runtime** — a proposal is not a selection until this returns:\n\n```bash\n\"$DOFLOW\" classify --task-class \"\u003Cproposed>\" --calling-skill do-flow --json\n```\nBranch on the returned `outcome` field, not the exit code.\n- **`ACCEPTED`** — the returned `workflow` is this run's plan of record; read `stages`, `gates` and `handoff` off it rather than from memory.\n- **`REJECTED`** — **stop.** Print `message` verbatim (it already names `validClasses` and any `suggestions`), ask the user to choose from `validClasses`, then re-validate. Never substitute `feature`.\n  A rejection may be about **you** rather than the class (`reason: caller-not-a-stage`). Then the fix is to propose one of the classes in `fit.hostingClasses`, or to hand the work to the skill this class names for the stage you meant — not to re-propose the same class.\n- **Exit 2** — surface the message verbatim and stop.\n\nEvery stage this run enters comes from the accepted `workflow`'s `stages`; a phase the workflow does\nnot declare is not run here, however familiar it is from the `feature` chain.\n\n4. **Announce the selection** — one line before anything runs:\n   `\u003CtaskClass> → \u003Cworkflow.name>: \u003CstageIds joined by →>`, plus the gate ids if `gates` is\n   non-empty. This is the user's chance to correct a misclassification cheaply.\n\n5. **Pick the starting stage.**\n   - `feature`: resume from the first missing chain artifact — `!has_requirement` → the discovery\n     stage; `has_requirement && !has_design` → design; `has_design && !has_plan` → planning; all\n     three present → go straight to the pre-implementation gate in step 7.\n   - Every other class: start at the first entry in `stages`.\n   - `--from \u003Cphase>` overrides the above to deliberately re-run a stage. Match the phase to the\n     stage whose `skill` is that phase's skill (`brainstorm` → `do-brainstorm`, `design` →\n     `do-design`, `plan` → `do-plan`, `implement` → the stage with `mutatesSource: true`, `test` →\n     `do-test`, `review` → `do-code-review`). No stage matches in this workflow → stop and list the\n     stage ids it does have. Two stages match — `bug` and `refactor` each run `do-test` twice — →\n     ask which one.\n\n6. **Run the stages in order**, invoking each stage's `skill` Behavioral Flow directly.\n   - `optional: true` — decide from that stage's own `purpose`, then say which way you went and\n     why. Never drop an optional stage silently.\n   - `readinessTemplate` is non-null — consult readiness before entering the stage:\n     `\"$DOFLOW\" readiness --task-class \"\u003CtaskClass>\" --task-id \"\u003Ctask id>\" --mode workflow --json`.\n     Both `--task-class` and `--task-id` are required; omitting either exits 2 and names the valid\n     set. Act on the returned `stageEntry.decision` — the entry policy is the runtime's, not this\n     page's: `ENTER` → enter the stage; `GATHER_FIRST` → report the missing items and gather them\n     first, do not enter; `ASK_USER` → ask the owed decision and wait; `STOP` → stop and surface\n     the conflicted claim.\n   - `readinessTemplate` is `null` — enter the stage. Do not consult readiness \"to be safe\": a\n     workflow whose `requiresImplementationReadiness` is `false` has no implementation to be ready\n     for and declares no template by design (`references\u002Ftask_classes.md`). Calling it anyway\n     invents a gate the registry does not declare.\n   - After each stage, report one line: stage id, skill, and the artifact path or result it\n     produced — so the user can follow without intervening.\n\n7. **Stop at the gates the resolved workflow declares** — read them from `workflow.gates`, each\n   attached to a stage by `afterStage`. `trigger: always` stops every time; any other trigger stops\n   only when that named condition actually holds. Ask with the gate's own `prompt` via\n   the question tool. For the `feature` workflow's three, which behave as they always have:\n   - **`gate-0`** (after discovery, trigger `unresolved-clarifications`) — a safety net, not the\n     normal path: `do-brainstorm` resolves ambiguities to zero via its own loop, so this fires only\n     if a `[NEEDS CLARIFICATION]` marker survived an aborted session. Ask one question per marker\n     (grouped if closely related), patch the answers into `requirement.md` directly, and do not\n     re-run `do-brainstorm`. Then follow steps 1-4 of the decision step in the guidance\n     tree's `references\u002FWORKFLOW_HANDOFF.md` (stage `discovery`, channel `question`) so the answers\n     are registered and cited; do-brainstorm already recorded discovery's handoff, so record none.\n   - **`gate-a`** (after planning) — the conversational checkpoint on top of the already-existing\n     advisory hook (`pre-implement-gate.sh`; its scope is in `references\u002FDOFLOW_CHAIN.md`), which\n     still reports a missing artifact. Options: `Proceed` \u002F `Let me review the artifacts first` \u002F `Stop here`.\n   - **`gate-b`** (after review) — no enforcement hook exists for this today; `\u002Fdo-git` is a\n     separate, always-manually-invoked skill. Frame the question by the review's own approval\n     status: on `CHANGES REQUESTED`, foreground addressing the findings rather than offering a bare\n     proceed.\n   A workflow with an empty `gates` array stops at none of these. It still stops on a readiness\n   block, on an ambiguous answer, and on any stage skill's own stopping rule.\n\n   **Record every answer that is a decision.** Asking is not recording: a gate this skill asked but\n   never wrote back stays open in the run, so the stage skill that owns it asks the same question\n   again — the user answers twice for one decision. Immediately after the user answers, write it:\n   ```bash\n   \"$DOFLOW\" orchestrate --action decide-gate --task-id \"\u003Cslug>\" --gate \"\u003Cgate id>\" --decision approve|reject --note \"\u003Cthe user's own answer, one line>\" --json\n   ```\n   Map each gate's own option set onto the two decisions the runtime accepts, and never onto a third:\n   - **`gate-a`** — `Proceed` is `approve`; `Stop here` is `reject`. `Let me review the artifacts\n     first` is **neither**: it is a pause, not a decision, so call nothing and actually pause — hold\n     the run at this gate and re-ask once the user comes back, exactly as step 8's ambiguous-answer\n     rule already requires.\n   - **`gate-b`** — an actual \"proceed to commit\u002Fmerge\" answer is `approve`; an explicit stop is\n     `reject`. Anything that is not a real decision — \"let me address the findings first\" on a\n     `CHANGES REQUESTED` review, most often — is neither, and gets no `decide-gate` call. `reject`\n     terminates the run outright, which is a different answer from \"not now\".\n   - **`gate-0`** — only in the aborted-session branch above, where this skill patched\n     `requirement.md` itself: once every surviving marker is resolved, `approve`. If any marker is\n     still open, leave the gate alone.\n   This call is advisory to the trail, not to the chain: if it fails for a reason outside this flow's\n   control (an unwritable local state directory, say), report the failure plainly and carry the\n   user's answer forward anyway — the answer is what governs whether the next stage runs, and the\n   record of it is what keeps the next stage from asking again.\n\n   When the answer also carries a choice beyond approve or reject and `paths --json` reports\n   `has_decisions: true`, register it too, once `decision --action list`\n   shows no live decision with that statement; the receiving stage's batch then leaves it out:\n   ```bash\n   \"$DOFLOW\" decision --action add --topic \"\u003Ckey>\" --statement \"\u003Ctext>\" --channel gate --stage \"\u003Cstage the gate follows>\" --rationale \"\u003Ctext>\" --json\n   ```\n\n8. **Never skip a gate on an ambiguous answer** — an unanswered or unclear response to the\n   question tool means stop and ask again, per RULE_04 in `rules\u002FRULES.md`. When the\n   terminal stage completes, state `workflow.handoff` as the next step rather than continuing into\n   it.\n\n## Conditional research within a feature\n\nBefore a feature stage relies on an external fact that local or user evidence cannot establish, or when the developer explicitly asks a research question, let that stage follow `..\u002Fdo-document\u002Freferences\u002Fembedded-research.md`. The stage opens a same-task `research-request`, records a source-linked finding or gap, then resumes itself without adding a seventh workflow stage. On resume, read request status and context pack. A blocking unanswered question prevents stage completion; neither research nor a retrieved page approves any existing gate. Feature discovery and design run the ideation pass; otherwise search only for a gap or an explicit request.\n\n## Boundaries\n**Will:**\n- Propose a task class, validate it through the runtime, and run the resolved workflow's stages in\n  order without a manual re-invocation at each boundary.\n- Stop only where the resolved workflow says to: its declared gates, a readiness block, or an\n  ambiguous answer.\n- Record each gate answer it collects through `orchestrate --action decide-gate`, so the stage skill\n  that owns that gate finds it already resolved instead of asking the user the same question twice.\n- Compose with the existing `pre-implement-gate.sh` advisory hook rather than bypass or duplicate it.\n  That hook keys on branch and artifact state, not on class, so it can advise against a `bug` or\n  `trivial-edit` workflow's edits on a branch that already holds an incomplete feature dir —\n  surface its message as written.\n\n**Will Not:**\n- Run any workflow for a class the runtime rejected, or replace a rejected class with `feature`.\n- Infer a class silently when two fit, or read a class the user stated as a suggestion.\n- Consult implementation readiness for a stage that declares no readiness template, or invent a\n  readiness template for a class that has none.\n- Modify any stage skill, or invoke `do-constitution` — it is standalone and in no class.\n- Add a hook-level enforcement gate for commit\u002Fmerge — `gate-b` is conversational only. A\n  hook-level version would be a separate, future proposal.\n- Silently proceed past a gate on an ambiguous or missing answer.\n- Record a pause or a deferral as a gate decision — `Let me review the artifacts first` and \"let me\n  address the findings first\" are neither `approve` nor `reject`, and forcing them into one writes a\n  decision the user did not make.\n\n**Next Step:** whatever the resolved `workflow.handoff` names — for every class that ends in a\nchange, `\u002Fdo-git` to commit\u002Fmerge once the last gate is cleared.\n",{"name":5,"effort":24,"description":7,"argument-hint":25},"high","[feature description] [--from brainstorm|design|plan|implement|test|review]","main","1f1bb66cc55e4d4f3d648089d35ea1f3826ce40a","core\u002Fshared\u002Fskills\u002Fdo-flow","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow","https:\u002F\u002Fgithub.com\u002Fkhoavu882","2026-11-10 18:44:15",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add khoavu882\u002Fdo-flow --skill do-flow",{"tool":38,"command":39},"bunx","bunx skills add khoavu882\u002Fdo-flow --skill do-flow",{"tool":41,"command":42},"pnpm","pnpm dlx skills add khoavu882\u002Fdo-flow --skill do-flow","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo-flow\u002FSKILL.md",true,[46],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Ftask_classes.md","REFERENCE","参考文档",8881,"https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo-flow\u002Freferences\u002Ftask_classes.md",[53,59,65,72,79,86],{"skillNo":54,"name":55,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M78BWJFR71W5PFWF9JZFQZD2","beta","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fbench-noise\u002Fskills\u002Fbeta\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M78BWGHVFZRKM7S4JYGPRA3Y","alpha","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fbench-noise\u002Fskills\u002Falpha\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M78BWB5C33WTE4YTSTBWJ01C","do","DoFlow universal dispatcher and the entry point for a plain code-change request: it classifies the task, routes it to the matching \u002Fdo-* workflow and applies that workflow's gates. Use whenever the user asks to add, fix, change or build something in code without naming a \u002Fdo-* skill, a request bundles several unrelated asks, or needs a tool recommendation or effort estimate first (e.g., 'add a retry to the upload client in src\u002Fupload.js', 'give me a rough estimate for this change', 'first do X, then document Y'). Skip it for questions about how existing code behaves, explanations, reviews, and work already inside a \u002Fdo-* skill.","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M78BW76N1YR7QRVXBQAV3QBN","do-test","Execute project builds, automated test suites, and coverage verification with intelligent error reporting across whatever toolchain the repo already uses. Use whenever the user wants to run existing tests, execute test suites, verify builds, check test coverage, or check compile status (e.g., 'run the tests', 'execute test suite', 'check coverage', 'verify build', 'does this still build'). Always activate this skill when verifying test suites or running builds without writing new application code.","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo-test\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M78BW2RW6HSN8D43AM83P1AV","do-plan","Generate the implementation plan (HOW) and dependency-ordered task checklist from requirement.md + design.md (plus design\u002Fspecs.md's IC-### interface contracts when the feature has one), with a Constitution Check gate, as Phase 3 of the doflow chain. Use whenever requirement.md and design.md exist and the next step is creating an actionable, owner-and-file-scoped task breakdown before implementation, or when the user says 'create an implementation plan', 'break this down into tasks', 'plan the work', or 'plan Phase 3'. Always activate this skill before executing feature implementation.","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo-plan\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M78BVXDWJ66PYCWNZR706GHE","do-implement","Direct, standalone code implementation from a description, \u002Fdo-code-review findings, or an existing task — no DoFlow chain artifacts required. Use when the user names \u002Fdo-implement, says 'now implement that', 'fix what the review found' or 'address these comments', or \u002Fdo routed a bug, refactor, dependency change or small edit here. A plain change request that names no \u002Fdo-* skill goes to \u002Fdo first, which classifies it and routes it back here when it fits. Skip it for questions, reviews, and new features that need the full chain.","https:\u002F\u002Fgithub.com\u002Fkhoavu882\u002Fdo-flow\u002Fblob\u002Fmain\u002Fcore\u002Fshared\u002Fskills\u002Fdo-implement\u002FSKILL.md",[],[]]