git:20260917.5a221da to git:20260917.f5ab5b1

---

name: dough-execute-plan

description: >-

Executes one selected story or bounded retrospective correction through an

executable plan, or one authorized planless slice from a selected simple story

or a contextual instruction, with independent refactoring, delivery, and

asynchronous CI repair. Use to execute a plan, run slices, execute a canonical

story when the caller explicitly skips slice planning, or execute a small

instruction from context without a story or plan. Does not decide story scope

or quick-path eligibility. `--trunk` selects Trunk Mode; omitted mode keeps

Story Branch Mode. `--skip-retro` skips the automatic planned-execution

retrospective. `--replan` and `--no-replan` choose whether an oversized attempt

may continue through planning.

---

# Execute planned or planless work

Execute the selected work through its existing plan. An explicit current instruction

may instead authorize one planless quick slice: an understood canonical feature story,

or a contextual instruction with no story or plan. The coordinator owns delivery;

implementation agents return uncommitted changes.

## Establish execution context

Identify the execution source before changing project state:

- **Planned:** require an executable plan. For a feature story, read its seed section.

For a bounded correction, require the complete correction input in the plan under

[planning scope and lifecycle](../dough-story-refinement/references/planning.md#choose-the-planning-level);

do not require or create a seed.

- **Quick:** require an explicit current instruction that authorizes planless

execution. The source is either a canonical feature story with understood goal,

scope, key examples, and no blocking decision, plus the instruction to skip slice

planning; or a contextual instruction that supplies the objective, known

expectations, remaining uncertainty, and authority. A seed alone or apparent story

size cannot select this path. Corrections require plans. Do not fabricate a story,

plan, or queue entry when the instruction is otherwise sufficient.

Name missing authority or source context and stop before backlog changes, delegation,

or implementation. Successful quick execution keeps scope, decisions, progress, and

proof in the conversation; create no plan, completion note, or substitute record.

A contextual instruction may leave expectations unresolved. Carry its goal and

uncertainty, resolve what a behavior change requires, and allow an evidenced

no-change conclusion through the explained-empty-change path.

Use [planning level](../dough-story-refinement/references/planning.md#choose-the-planning-level)

for source ownership, [proof ownership](../dough-story-refinement/references/planning.md#own-executable-proof)

when mapping or accepting proof, and [active-plan refinement](../dough-story-refinement/references/planning.md#refine-the-active-plan)

for plan updates. Retain completed plans, source history, and review evidence for

retrospective and [story wrap-up](../dough-story-wrap-up/SKILL.md); quick work retains

its source, conversation, and execution identity. Use [slice decomposition](../dough-story-decomposition/references/problem-decomposition.md#decompose-slices)

for Behavior/Structure types and [slice sizing](../dough-story-decomposition/references/problem-decomposition.md#size-and-escalate-slices)

for budgets and learning escalation.

At startup, obtain enough authoritative context to select and delegate the first slice.

Reuse instructions while their sources and applicability assumptions remain unchanged.

Resolve project context at the first boundary that needs it:

- execution-source kind, slice target, hard limit, and exceptions;

[replanning permission](references/execution-decisions.md#choose-replanning-permission);

execution mode and location: default Story Branch Mode; `--trunk` or a clear

equivalent selects Trunk Mode; explicit caller selection uses the current

branch. Resolve contradictions before changing state. Mode never creates

execution authority; for planned work, plan path and status vocabulary;

- backlog path and selected entry for work selected from **Backlog list**;

- selective formatter and commit hook contract before taking queued work and its claim commit;

- navigation, focused tests, runtime wrapper, and workflow precedence for the selected slice;

- authorized push destination before delivery, and for Trunk Mode before

publishing a queue claim or verified increment;

- generation triggers and commands when affected; and

- [refactor context](../dough-post-change-refactor/SKILL.md) before refactor delegation.

Missing context stops its affected boundary, including first-slice delegation when

needed there. Other invoking tools, including GSD, retain this slice delivery contract;

a phase or task cannot replace a slice.

Before first implementation delegation, read [delegation](references/delegation.md)

and the common reassessment/human-judgment rules, [replanning

permission](references/execution-decisions.md#choose-replanning-permission), plus

currently triggered sections of [execution decisions](references/execution-decisions.md).

Before accepting a return, read [proof acceptance](references/wrap-up.md#accept-proof);

before delivery, read [delivery](references/wrap-up.md#deliver-the-change). Before

arming observation, read [CI observation](references/ci-monitor.md) and only the

current host's notification adapter. Arm from the execution checkout against the

authorized target branch; do not wait for CI. Before creating the execution workspace, read

[execution location](references/execution-location.md). Before a Trunk Mode

claim or increment publication, read [trunk publication](references/trunk-publication.md).

Use [targeted retrieval and disposable research](references/disposable-research.md)

for omitted/truncated passages or bounded investigations; another step alone needs no reload.

## Take queued work

If an authorized Git operation reports conflicts, inspect the unmerged paths

for this project's product backlog (often `PRODUCT-BACKLOG.md`). Before editing

or staging its resolution, read and apply

[backlog merge conflicts](../dough-product-backlog/references/merge-conflicts.md),

including its staged-result verification before continuing the Git operation.

If the reference is unavailable, preserve the conflict and report the missing guidance.

After resolving execution source and authority, inspect the backlog before plan-status

changes, observer recovery/startup, delegation, or implementation. Moving a selected

Before moving it, resolve selective formatting and the Taken-only commit's hook contract.

An absent or understood check-only hook permits the transition. An unknown, mutating,

failing, or disputed hook stops it with the queue unchanged until safely resolved through

execution decisions. Resolution runs neither delivery formatting nor hook-owned lint;

obtain push/CI context only when another current boundary needs it. Then follow

[take queued work](../dough-product-backlog/SKILL.md#take-queued-work-for-execution).

An ambiguous move stops implementation.

Already **Taken** means resume: preserve its position without duplication. Work absent

from both active lists needs no fabricated entry. Planning/refinement never takes work.

Leave taken work through pauses, failures, completion, and retrospective; wrap-up removes it.

After moving, stage only the backlog path, inspect the staged diff, and commit the claim

complete it before dispatch. Staging/commit failure stops isolated execution:

preserve and report backlog/index state. No-change cases produce no empty claim

commit.

Story Branch Mode creates the execution branch/worktree from that local commit

only after success; do not push the claim separately. Trunk Mode publishes the

claim per [trunk publication](references/trunk-publication.md#publish-a-queue-claim)

before implementation. Later workspace-setup failure leaves the published or

locally committed **Taken** entry for retry; do not treat that as a new claim.

## Choose the execution location

Follow [execution location](references/execution-location.md) for mode,

workspace creation, retained identity, resume, push destination, and

checkout-bound runtime.

## Continue or recover at an execution boundary

Planned work uses the existing plan and conversation under

[execution and resume state](../dough-story-refinement/references/planning.md#write-an-executable-plan).

Quick work uses its source and conversation, without a recovery artifact. Retain the

resolved replanning permission with that context. During uninterrupted

work, reuse decisions, accepted proof, and delivery progress while their sources,

assumptions, and covered boundaries hold. After confirmed push, obtain only newly relevant

next-slice detail; a slice transition alone needs no full recovery read.

After interruption or a changed observation from Git, agents, or observer coverage, verify

execution identity first and reconcile only affected worktree/index, branch/commits,

implementation/refactor return, and exact observer identity. Preserve unrelated or ambiguously

owned work. Reuse proof only while promise, boundary, implementation, setup, and observations match.

Resume at the first delivery obligation not established by evidence. An incomplete

or oversized return still needs [oversized-slice handling](references/execution-decisions.md#refine-an-oversized-slice)

before proof acceptance. Otherwise implementation returns still need proof

acceptance/refactoring; completed refactors need remaining delivery;

uncommitted plan edits need staging/commit. Classify a Trunk Mode increment with

[interrupted publication](references/trunk-publication.md#resume-an-interrupted-publication)

before any further commit or push. Story Branch Mode still pushes local commits

absent from its authorized destination. Plan status or a compact report proves none

of those later boundaries. When pushed commit, retained delivery result, and

required registration agree, select the next dependency-ready slice.

Missing/contradictory execution identity requires the recovery decision above.

## Execute the next slice

1. In the execution checkout, use the plan's current statuses, decisions, learnings,

proof, and selected existing-solution finding/evidence when present. Read retained state

on initial entry/recovery; reuse valid readings during continuation. For quick work,

reread the source and conversation's scope, decisions, progress, proof, and remaining

uncertainty; it is the only slice, with no separate state artifact. Confirm current

execution authority. Recover an existing CI observer before considering a new one.

2. Select the next unfinished dependency-ready planned slice, or the one quick slice. Apply

[execution decisions](references/execution-decisions.md); for behavior/state removal or

disablement, also run the [destructive later-outcome check](references/destructive-later-outcome-check.md).

3. When planned refinement is needed and learning escalation permits, invoke

[slice-plan refinement](../dough-slice-plan-refinement/SKILL.md) in place, then restart

at step 1, unless replanning is disabled; then apply

[oversized-slice decisions](references/execution-decisions.md#refine-an-oversized-slice)

and stop without retry. If quick work no longer fits one coherent slice, apply

[oversized-slice decisions](references/execution-decisions.md#refine-an-oversized-slice).

A no-replan return stops without planning or retry. When replanning is allowed, use

[ordinary slice planning](../dough-slice-planning/SKILL.md) for remaining work,

and restart as planned execution. Before

delegating a change that invalidates a required

pre-change observation, apply [proof ownership](../dough-story-refinement/references/planning.md#own-executable-proof):

reuse an adequate baseline with known matching revision/environment/selection conditions,

or obtain it first. Missing/failed prerequisites stop only dependent work. Apply on entry

and resume without a new startup audit or repeated recovery read. Otherwise delegate

under [delegation](references/delegation.md).

4. On return, recheck execution decisions; handle incomplete/oversized work there before

delivery, including a no-replan overrun. Otherwise [accept proof](references/wrap-up.md#accept-proof) and confirm

uncommitted work or an explained empty change.

5. Run [delivery](references/wrap-up.md#deliver-the-change) end to end. After

successful delivery, restart for remaining planned slices; a delivered

quick slice has no successor.

Planned slices may run concurrently only with disjoint file changes, mutable state, and

plan writes. Quick execution has one slice. Each slice completes coordinator-owned delivery

before a dependent slice starts.

## Finish or stop

On completion, human-judgment stop, or cancellation, close the observer through the current

host adapter: handle delivered failures, then stop observers without waiting for CI. Report pending CI

as unobserved.

After all planned slices satisfy proof/delivery and required observer shutdown succeeds,

report completion, retained evidence, and CI limitations. `--skip-retro` skips only this

execution's automatic retrospective; it changes no preferences or proof/delivery/shutdown

obligations. Explicit omit/defer instructions also take precedence. When skipped, report it

and end with `## PLAN EXECUTION COMPLETE`, retaining plan/evidence for later review and wrap-up.

Otherwise report `## PLAN EXECUTION COMPLETE`, then invoke

[dough-execution-retrospective](../dough-execution-retrospective/SKILL.md) without another

confirmation. Preserve explicit review instructions and project preferences through its

review selection; its authority excludes implementing findings or changing the backlog.

Continue in the recorded execution project/checkout. Supply available references/context:

source contract, original plan and approved changes, attributable commits, decisions, proof,

delivery state, CI limitations, and checkout/branch identity. Trunk Mode attributable

commits are that identity's retained published revisions, not another ledger or a

rewrite's unpublished SHA. Include an initial quick attempt and its planned

continuation as one execution. Reuse context without another handoff artifact

or transcript copy; retrospective validates attribution and recovers real gaps.

Execution completion and review completion are distinct. A retrospective context stop leaves

execution complete: report missing input without rerunning implementation or claiming review

completion. On recovery, use retained review state to continue or recognize completed review;

ambiguous state requires recovery rather than duplicate review or guessed completion.

Retain the completed plan, evidence, execution checkout, branch, and worktree for story

wrap-up; do not invoke it here. Wholly planless completion retains source, conversation,

identity, delivered changes, and proof, reports delivered work and shutdown, and ends with

`## QUICK EXECUTION COMPLETE` after required delivery/shutdown, without automatic

retrospective. The coordinator invokes wrap-up after successful branch delivery. Do not

report integrated completion here; wrap-up owns Story Branch merge and required

target push, Trunk Mode closure publication, and resource cleanup.

For incomplete work, failed delivery/shutdown, cancellation, or a human-judgment stop, report

source, active plan/next slice or quick-slice state, preserved work, observer state,

and required decision/recovery action. Emit no completion marker or automatic retrospective.