Subagent-Driven Development

Execute plan by dispatching a fresh implementer subagent per task, a task review (spec compliance + code quality) after each, and a broad whole-branch review at the end.

Why subagents: You delegate tasks to specialized agents with isolated context. By precisely crafting their instructions and context, you ensure they stay focused and succeed at their task. They should never inherit your session's context or history — you construct exactly what they need. This also preserves your own context for coordination work.

Core principle: Fresh subagent per task + task review (spec + quality) + broad final review = high quality, fast iteration

Narration: between tool calls, narrate at most one short line — the ledger and the tool results carry the record.

Continuous execution: Do not pause to check in with your human partner between tasks. Execute all tasks from the plan without stopping. The only reasons to stop are the four named below, or all tasks complete. "Should I continue?" prompts and progress summaries waste their time — they asked you to execute the plan, so execute it.

Rulings, not stalls. A running plan does not wait on a human. Conflicts, ambiguities, plan defects, a cap you would have asked to exceed — decide them. The spec is the binding authority, the plan is its argument, and your judgment settles what neither answers. Record every decision in the ledger as Ruling: <what you decided> — <why> — <what it costs if wrong>, and keep going. A wrong ruling costs rework your human partner can see and undo; a session parked on a question costs their whole day and buys nothing.

Four things stop you, and only these: an irreversible or destructive operation; a security-sensitive action; a side effect outside this worktree that norms say you ask about first (a merge, a push to a shared branch, a publish); and a plan so broken that every path forward is a guess. For those, stop and ask.

When to Use

digraph when_to_use {
    "Have implementation plan?" [shape=diamond];
    "Tasks mostly independent?" [shape=diamond];
    "Partner chose inline, or no subagent tool?" [shape=diamond];
    "sdd-subagent-driven-development" [shape=box];
    "sdd-executing-plans" [shape=box];
    "Manual execution or brainstorm first" [shape=box];

    "Have implementation plan?" -> "Tasks mostly independent?" [label="yes"];
    "Have implementation plan?" -> "Manual execution or brainstorm first" [label="no"];
    "Tasks mostly independent?" -> "Partner chose inline, or no subagent tool?" [label="yes"];
    "Tasks mostly independent?" -> "Manual execution or brainstorm first" [label="no - tightly coupled"];
    "Partner chose inline, or no subagent tool?" -> "sdd-executing-plans" [label="yes"];
    "Partner chose inline, or no subagent tool?" -> "sdd-subagent-driven-development" [label="no"];
}

vs. Executing Plans (inline):

Fresh subagent per task (no context pollution) instead of one context doing every task
Review after each task (spec compliance + code quality) instead of only at the end
Costs a fresh context per task and per review; inline costs one context plus one final reviewer
Both run in this session, share the same plan workspace and ledger, and never pause between tasks

The Process

digraph process {
    rankdir=TB;

    subgraph cluster_per_task {
        label="Per Task";
        "Dispatch sdd-implementer (./implementer-prompt.md)" [shape=box];
        "Implementer asks questions?" [shape=diamond];
        "Answer questions, provide context" [shape=box];
        "Implementer implements, tests, commits, self-reviews" [shape=box];
        "Generate review package, dispatch sdd-task-reviewer (./task-reviewer-prompt.md)" [shape=box];
        "Spec ✅ and quality approved?" [shape=diamond];
        "Finding conflicts with plan text?" [shape=diamond];
        "Rule on the conflict, ledger the ruling" [shape=box];
        "Fix round R of 5: R≤3 resume implementer; R≥4 sdd-implementer-standard" [shape=box];
        "Dispatch sdd-task-reviewer (./re-review-prompt.md)" [shape=box];
        "All findings addressed?" [shape=diamond];
        "R = 5?" [shape=diamond];
        "Adjudicate each open finding" [shape=box];
        "Any load-bearing finding?" [shape=diamond];
        "Rule and continue; stop only if every path forward is a guess" [shape=box];
        "Park findings in ledger with rulings" [shape=box];
        "Append completion to ledger, mark todo complete" [shape=box];
    }

    "Setup: worktree, ledger check, read plan, pre-flight review" [shape=box];
    "More tasks remain?" [shape=diamond];
    "Spec acceptance: run each Success Criterion, ledger results" [shape=box];
    "Dispatch sdd-final-reviewer (../sdd-requesting-code-review/code-reviewer.md)" [shape=box];
    "Final findings? ONE fix dispatch, one scoped re-review, adjudicate residuals" [shape=box];
    "Final review clean: delete this plan's workspace" [shape=box];
    "Use sdd-finishing-a-development-branch" [shape=box style=filled fillcolor=lightgreen];

    "Setup: worktree, ledger check, read plan, pre-flight review" -> "Dispatch sdd-implementer (./implementer-prompt.md)";
    "Dispatch sdd-implementer (./implementer-prompt.md)" -> "Implementer asks questions?";
    "Implementer asks questions?" -> "Answer questions, provide context" [label="yes"];
    "Answer questions, provide context" -> "Implementer implements, tests, commits, self-reviews";
    "Implementer asks questions?" -> "Implementer implements, tests, commits, self-reviews" [label="no"];
    "Implementer implements, tests, commits, self-reviews" -> "Generate review package, dispatch sdd-task-reviewer (./task-reviewer-prompt.md)";
    "Generate review package, dispatch sdd-task-reviewer (./task-reviewer-prompt.md)" -> "Spec ✅ and quality approved?";
    "Spec ✅ and quality approved?" -> "Append completion to ledger, mark todo complete" [label="yes"];
    "Spec ✅ and quality approved?" -> "Finding conflicts with plan text?" [label="no"];
    "Finding conflicts with plan text?" -> "Rule on the conflict, ledger the ruling" [label="yes"];
    "Rule on the conflict, ledger the ruling" -> "Fix round R of 5: R≤3 resume implementer; R≥4 sdd-implementer-standard";
    "Finding conflicts with plan text?" -> "Fix round R of 5: R≤3 resume implementer; R≥4 sdd-implementer-standard" [label="no"];
    "Fix round R of 5: R≤3 resume implementer; R≥4 sdd-implementer-standard" -> "Dispatch sdd-task-reviewer (./re-review-prompt.md)";
    "Dispatch sdd-task-reviewer (./re-review-prompt.md)" -> "All findings addressed?";
    "All findings addressed?" -> "Append completion to ledger, mark todo complete" [label="yes"];
    "All findings addressed?" -> "R = 5?" [label="no"];
    "R = 5?" -> "Fix round R of 5: R≤3 resume implementer; R≥4 sdd-implementer-standard" [label="no - next round"];
    "R = 5?" -> "Adjudicate each open finding" [label="yes - breaker trips"];
    "Adjudicate each open finding" -> "Any load-bearing finding?";
    "Any load-bearing finding?" -> "Rule and continue; stop only if every path forward is a guess" [label="yes"];
    "Any load-bearing finding?" -> "Park findings in ledger with rulings" [label="no"];
    "Park findings in ledger with rulings" -> "Append completion to ledger, mark todo complete";
    "Append completion to ledger, mark todo complete" -> "More tasks remain?";
    "More tasks remain?" -> "Dispatch sdd-implementer (./implementer-prompt.md)" [label="yes"];
    "More tasks remain?" -> "Spec acceptance: run each Success Criterion, ledger results" [label="no"];
    "Spec acceptance: run each Success Criterion, ledger results" -> "Dispatch sdd-final-reviewer (../sdd-requesting-code-review/code-reviewer.md)";
    "Dispatch sdd-final-reviewer (../sdd-requesting-code-review/code-reviewer.md)" -> "Final findings? ONE fix dispatch, one scoped re-review, adjudicate residuals";
    "Final findings? ONE fix dispatch, one scoped re-review, adjudicate residuals" -> "Final review clean: delete this plan's workspace";
    "Final review clean: delete this plan's workspace" -> "Use sdd-finishing-a-development-branch";
}

Setup

Ensure the work happens in an isolated workspace: use sdd-using-git-worktrees to create one or verify the existing one. Never start implementation on a main/master branch without your human partner's explicit consent.

Scripts. SKILL_DIR below is the directory containing this SKILL.md. Resolve it once and invoke its scripts by absolute path — bash "$SKILL_DIR/scripts/task-brief" ... — while keeping your CWD at the repo root, so relative plan paths and git resolve. (bash scripts/... only works with CWD set to SKILL_DIR, which breaks project-relative paths; a bare relative path from the repo root does not exist.)

Conversation memory does not survive compaction. In real sessions, controllers that lost their place have re-dispatched entire completed task sequences — the single most expensive failure observed. Track progress in a ledger file, not only in todos.

Each plan owns a workspace: at skill start, run this skill's bash "$SKILL_DIR/scripts/sdd-workspace" PLAN_FILE — it prints the plan's git-ignored directory (under <repo-root>/.superpowers/sdd/), home to every artifact for THIS plan: ledger, briefs, reports, review packages. Another plan's directory is never yours to read or write.
Check for this plan's ledger at <workspace>/progress.md. If its first line names your plan file, tasks with a Task <N>: complete line are DONE — do not re-dispatch them; resume at the first task without one. A task whose last line is a fix round is mid-loop: resume the loop at the next round. A ledger whose first line names a different plan file — or a stray ledger at the old flat path .superpowers/sdd/progress.md — is another plan's progress: leave it in place and start your own, fresh.
Create the ledger with its identity as the first line: # SDD ledger — plan: <plan file path>.
The ledger is your recovery map: the commits it names exist in git even when your context no longer remembers creating them. After compaction, trust the ledger and git log over your own recollection.
git clean -fdx will destroy the workspace (it's git-ignored scratch); if that happens, recover from git log.

Read the plan once, note its context and Global Constraints, and create a todo per task. If the plan names a Spec, read that too: the spec is the authority the plan argues from, and conflicts inside the plan resolve against it. A plan with no reachable spec gets a ledger note saying so — rulings made without one are provisional.

Before dispatching Task 1, scan the plan once for conflicts, writing down what you checked as you check it:

tasks that contradict each other or the plan's Global Constraints
anything the plan explicitly mandates that the review rubric treats as a defect (a test that asserts nothing, verbatim duplication of a logic block)

The scan's output is a table, not a verdict. One row for every pair of tasks that share a file or an interface: the two tasks, what one produces against what the other consumes, and what you found. One row for every task: whether its own text agrees with itself — the tests it specifies against the code it specifies, the files it creates against the files it later touches. "The scan is clean" without those rows is not a scan you ran.

Write the table to the ledger. Rule on everything you find before execution begins — each finding against the plan text that mandates it — and record each ruling in the ledger. If the scan is clean, proceed without comment. Rule on each conflict it surfaces — the spec is the binding authority, the plan is its argument — record the ruling beside its row, and dispatch Task 1. The review loop remains the net for conflicts that only emerge from implementation.

Agent Selection

Dispatch every seat as its named agent type. Each seat's definition pins the model, thinking variant, and step budget; a generic agent inherits the session's model and silently defeats this section. Use the least capable agent that can handle the role.

Implementer, per task: sdd-implementer for mechanical tasks (isolated functions, complete specs, 1-2 files). sdd-implementer-standard for integration and judgment (multi-file coordination, pattern matching, debugging) and for architecture-sensitive work.
Fix-loop escalation (rounds 4-5): sdd-implementer-standard — at least one tier above the implementer that got stuck.
Task review and scoped re-review: sdd-task-reviewer.
Final whole-branch review: sdd-final-reviewer — a judgment seat, never the session default.
Other seats: sdd-spec-reviewer (spec review, 01 DEFINE), sdd-plan-reviewer (plan review, 02 PLAN), sdd-designer (design-it-twice sketches, 02 PLAN), sdd-researcher (background research), sdd-falsifier (falsification pass), sdd-debugger (parallel independent failures).

Turn count beats token price. Wall-clock and context cost scale with how many turns a subagent takes, and the cheapest models routinely take 2-3× the turns on multi-step work — costing more overall. The seats are configured at the tiers this section describes. When the task's plan text contains the complete code to write, the implementation is transcription plus testing — sdd-implementer covers it; single-file mechanical fixes also take sdd-implementer.

Task complexity signals (implementation tasks):

Touches 1-2 files with a complete spec → sdd-implementer
Touches multiple files with integration concerns → sdd-implementer-standard
Requires design judgment or broad codebase understanding → sdd-implementer-standard

The Task Loop

Batch small same-shape work. When the plan lists several tasks that are each a small, independent edit of the same kind — the same one-line fix, constant change, or field addition repeated across files — do not dispatch one subagent per task. Run task-brief for every task in the batch and hand the one implementer all their brief paths; the batch shares one report file (name it …/batch-<first>-<last>-report.md) and one review package over the batch's BASE..HEAD range. Each batched task still gets its own Task <N>: complete (commits <base7>..<head7>, …) ledger line when the batch's review comes back clean, all sharing the range. Reserve one-dispatch-per-task for work that needs its own judgment, its own tests, or its own review surface.

Everything you paste into a dispatch prompt — and everything a subagent prints back — stays resident in your context for the rest of the session and is re-read on every later turn. Hand artifacts over as files.

Waiting on dispatched subagents: never poll a wait interface with short timeouts, and never sit in one silent, open-ended wait either. While you have local work — ledger updates, packaging the next review, reading reports — keep working; child results arrive on their own. When you are genuinely idle, wait in bounded stretches (five to ten minutes, where your platform allows), and between stretches post one line of status and reconcile your live children: list them, and chase any that finished without reporting. A bounded stretch keeps nearly all of a long wait's efficiency while guaranteeing a stuck or lost child is noticed within minutes, not at the end of the session.

1. Dispatch the implementer

Record BASE (git rev-parse HEAD) before dispatching — the review package and fix-round diffs need it.

Task brief: before dispatching an implementer, run this skill's bash "$SKILL_DIR/scripts/task-brief" PLAN_FILE N — it extracts the task's full text to a uniquely named file and prints the path. It also copies the plan header's Global Constraints ahead of the task — they bind every task, and the brief is the implementer's only view of the plan. Compose the dispatch so the brief stays the single source of requirements. Your dispatch should contain: (1) one line on where this task fits in the project; (2) the brief path, introduced as "read this first — it is your requirements, with the exact values to use verbatim"; (3) interfaces and decisions from earlier tasks that the brief cannot know; (4) your resolution of any ambiguity you noticed in the brief; (5) the report-file path and report contract. Exact values (numbers, magic strings, signatures, test cases) appear only in the brief. Never make a subagent read the whole plan file.
Report file: name the implementer's report file after the brief (brief …/task-N-brief.md → report …/task-N-report.md) and put it in the dispatch prompt. The implementer writes the full report there and returns only status, commits, a one-line test summary, and concerns.
A dispatch prompt describes one task, not the session's history. Do not paste accumulated prior-task summaries ("state after Tasks 1-3") into later dispatches — a real session's dispatch hit 42k chars of which 99% was pasted history. A fresh subagent needs its task, the interfaces it touches, and the global constraints. Nothing else.
The dispatch carries the no-subagents contract (it is in the implementer template): the implementer never dispatches subagents — not helpers, and never a reviewer. Review arrives from you, after the report. In real sessions, every reviewer a worker spawned duplicated the task review the controller dispatched anyway — a full extra review seat per task.
If an earlier task parked a finding in the area this task touches, carry a pointer to that ledger entry in the dispatch.
Record the implementer's agent identity from the dispatch result — fix-loop rounds 1-3 resume this agent.
Never dispatch multiple implementation subagents in parallel (conflicts).

Template: implementer-prompt.md

2. Handle the report

Implementer subagents report one of four statuses. Handle each appropriately:

DONE: Generate the review package (bash "$SKILL_DIR/scripts/review-package" PLAN_FILE BASE HEAD — it prints the unique file path it wrote; BASE is the commit you recorded before dispatching the implementer — never HEAD~1, which silently drops all but the last commit of a multi-commit task), then dispatch the task reviewer with the printed path. If the report shows tracked changes left uncommitted, the package omits them — send the implementer back to commit before packaging.

A "Noticed but not touching" list in a report is observations about work outside this task's scope: record any that are real defects as deferred minors in the ledger, and carry the rest to your final message. They never enter this task's fix loop.

DONE_WITH_CONCERNS: The implementer completed the work but flagged doubts. Read the concerns before proceeding. If the concerns are about correctness or scope, address them before review. If they're observations (e.g., "this file is getting large"), note them and proceed to review.

NEEDS_CONTEXT: The implementer needs information that wasn't provided. Provide the missing context and re-dispatch.

BLOCKED: The implementer cannot complete the task. Assess the blocker:

If it's a context problem, provide more context and re-dispatch to the same agent
If the task requires more reasoning, re-dispatch as sdd-implementer-standard
If the task is too large, break it into smaller pieces
If the plan itself is wrong, rule on the correction, ledger it, and re-dispatch with the ruling carried in the dispatch

Never ignore an escalation or force the same agent to retry without changes. If the implementer said it's stuck, something needs to change.

If the implementer asks questions — before starting or mid-task — answer clearly and completely, provide additional context if needed, and don't rush it into implementation.

3. Review the task

Per-task reviews are task-scoped gates. The broad review happens once, at the final whole-branch review. Never skip the task review, and never accept a report missing either verdict — spec compliance AND task quality are both required. Implementer self-review never replaces the task review; both are needed.

Hand the reviewer its diff as a file: run this skill's bash "$SKILL_DIR/scripts/review-package" PLAN_FILE BASE HEAD and pass the reviewer the file path it prints (or, without bash: git log --oneline, git diff --stat, and git diff -U10 for the range, redirected to one uniquely named file). The output never enters your own context, and the reviewer sees the commit list, stat summary, and full diff with context in one Read call. Use the BASE you recorded before dispatching the implementer — never HEAD~1, which silently truncates multi-commit tasks. Never dispatch a task reviewer without a diff file.
Reviewer inputs: the task reviewer gets three paths — the same brief file (every brief file for a batch), the report file, and the review package — plus the global constraints that bind the task.
review-package exit codes: 0 = package written; 2 = usage or a bad ref (fix the invocation); 3 = no package was written (range not rooted at BASE, empty commit range, or empty net diff). Exit 3 after a fix round means the fix was never committed — send it back to commit and re-run. Exit 3 on a fresh task range means the task left no commits: re-dispatch the implementer, or ledger a ruling that nothing was built. Never dispatch a reviewer without a package.
The global-constraints block you hand the reviewer is its attention lens. Copy the binding requirements verbatim from the plan's Global Constraints section or the spec: exact values, exact formats, and the stated relationships between components ("same layout as X", "matches Y"). The reviewer's template already carries the process rules (YAGNI, test hygiene, review method) — the constraints block is for what THIS project's spec demands.
Do not add open-ended directives like "check all uses" or "run race tests if useful" without a concrete, task-specific reason
Do not ask a reviewer to re-run tests the implementer already ran on the same code — the implementer's report carries the test evidence
Do not pre-judge findings for the reviewer — never instruct a reviewer to ignore or not flag a specific issue. If you believe a finding would be a false positive, let the reviewer raise it and adjudicate it in the review loop. If the prompt you are writing contains "do not flag," "don't treat X as a defect," "at most Minor," or "the plan chose" — stop: you are pre-judging, usually to spare yourself a review loop. The task reviewer may report "⚠️ Cannot verify from diff" items — requirements that live in unchanged code or span tasks. These do not block the rest of the review, but you must resolve each one yourself before marking the task complete: you hold the plan and cross-task context the reviewer lacks. If you confirm an item is a real gap, treat it as a failed spec review — it enters the fix loop with the other findings.

Template: task-reviewer-prompt.md

4. The fix loop

The loop triggers when the review reports spec ❌, any Critical or Important finding, or a ⚠️ item you confirmed as a real gap.

Before the loop starts, two routes leave it immediately:

Record Minor findings in the progress ledger as you go (Task <N>: minor (deferred): <one-liner>), and point the final whole-branch review at that list so it can triage which must be fixed before merge. A roll-up nobody reads is a silent discard. Minor findings never enter the loop.
A finding labeled plan-mandated — or any finding that conflicts with what the plan's text requires — is yours to rule on: weigh the finding against the plan text, decide with the spec as the binding authority, and ledger the ruling before you act on it. Do not dismiss the finding because the plan mandates it, and do not dispatch a fix that contradicts the plan without a recorded ruling. Everything else enters the loop. A fix round is one fix dispatch plus one scoped re-review. Five rounds maximum per task:

Rounds 1-3 — resume the original implementer. Send it the open findings verbatim. Its context is intact: it knows the task, the code, and its own choices. If your harness cannot send another message to a live subagent, dispatch a fresh implementer carrying the brief path, the report-file path, and the findings — the report file is the persistent memory either way.

Rounds 4-5 — dispatch a fresh implementer as sdd-implementer-standard (per Agent Selection), with the brief path, the report-file path, the open findings, and this framing: "A prior implementer attempted this task [N] times; you own it now. Read the report file for what was tried." A loop that survives three resumes usually means the implementer cannot see its own problem — fresh eyes and a capability bump in one move.

Every round, either way: the implementer fixes, re-runs the tests covering the amended code, appends its fix report to the same report file, and returns the short contract. Before re-dispatching the reviewer, confirm the fix report contains the covering tests, the command run, and the output; dispatch the re-review once all three are present. Name the covering test files in the fix message — a one-line fix does not need the whole suite.

The re-review is scoped. Run bash "$SKILL_DIR/scripts/review-package" PLAN_FILE FIX_BASE HEAD where FIX_BASE is the head the previous review saw (exit 3 here means the fix was never committed — apply §3's exit-code rule), and dispatch re-review-prompt.md with the findings list, the brief, the report file, and the printed diff path. The re-reviewer verdicts each finding ADDRESSED or NOT ADDRESSED and flags new breakage in the fix diff only. New Critical/Important breakage in the fix diff joins the open findings list. Out-of-scope observations go to the ledger as deferred minors — they never extend the loop.

After each round, append to the ledger: Task <N>: fix round <R>/5 (<X> addressed, <Y> open — <finding one-liners>; commits <a7>..<b7>)

Never fix findings yourself in the controller session — your context stays clean for coordination, and controller fixes skip review.

The breaker. When round 5's re-review still leaves findings open, stop dispatching. Adjudicate each open finding yourself — you hold the plan and the cross-task context the reviewer lacks:

The reviewer is wrong, or the point is contestable: park it — Task <N>: parked — <finding> — Ruling: <why the code stands>. The final review sees both sides.
Real, but nothing downstream builds on it: park it the same way, with a ruling that says it's real and deferred.
Real and load-bearing — a later task builds on it, or it reveals a plan defect: rule on the smallest change that unblocks the dependent work, ledger it as Task <N>: Ruling: <finding> — <what you decided and why>, and carry it into the next task's dispatch. Parking a structural failure silently lets every dependent task build on it. Stop only when the defect leaves every path forward a guess.

Adjudicate only at the cap. Adjudicating earlier to end a loop is pre-judging with a different name. Every adjudication is a ledger entry — a silent discard is forbidden.

5. Complete the task

When the review comes back clean — or every open finding is parked with a ruling at the cap — append the completion line to the ledger in the same message as your other bookkeeping:

Task <N>: complete (commits <base7>..<head7>, review clean)
Task <N>: complete (commits <base7>..<head7>, <K> parked) after a tripped breaker

Then mark the todo complete and move on. Never move to the next task while the review has open Critical/Important issues that are neither fixed nor parked-with-ruling at the cap.

Verify the Branch

Confirm the tree is clean first — git status --porcelain -uno prints nothing. Tracked changes left uncommitted are not in a ref-range package: commit them (or ledger a ruling discarding them) before running the criteria, or your evidence describes code the reviewer never saw.

Then run the spec acceptance pass no per-task gate covers. Re-read the spec's Confirmed Intent and Success Criteria and work each criterion:

Run its command or recorded procedure — from the spec's Success Criteria row or the plan's Commands block — against the branch, and ledger Final: spec acceptance — <criterion> → <command or procedure> → <result>.
A criterion with no runnable command or recorded procedure is a finding, not a pass — name the missing evidence source.
Confirm no shipped behavior crosses an Out of Scope item.

Findings here enter the final review's one fix wave. This is the branch-level seat; sdd-verification-before-completion governs the claims, and per-task TDD evidence does not stand in for it.

Final Review

The final whole-branch review gets a package too. Confirm git status --porcelain -uno is still empty, then run bash "$SKILL_DIR/scripts/review-package" PLAN_FILE MERGE_BASE HEAD (MERGE_BASE = the commit the branch started from, e.g. git merge-base main HEAD; exit 3 means no package — apply §3's exit-code rule) and include the printed path in the final review dispatch, so the final reviewer reads one file instead of re-deriving the branch diff with git commands. Dispatch as sdd-final-reviewer (see Agent Selection), using sdd-requesting-code-review's code-reviewer.md. Fill the template's inputs explicitly: [DESCRIPTION] (one line on what the branch built), [PLAN_OR_REQUIREMENTS] (the spec and plan paths), [BASE_SHA] (MERGE_BASE), [HEAD_SHA], and [DIFF_FILE] (the printed package path). Point it at the ledger's spec-acceptance, deferred-minor, and parked lines so it can triage which must be fixed before merge, and pass the plan's Review Focus section verbatim if it has one (the spec-silent cases the plan's tests pin — the reviewer checks each deliberately).

If the final whole-branch review returns findings, dispatch ONE fix subagent with the complete findings list — not one fixer per finding. Per-finding fixers each rebuild context and re-run suites; a real session's final-review fix wave cost more than all its tasks combined. Name a report file for it in the workspace (<workspace>/final-fix-report.md) and require the evidence the task loop asks for: what changed, the covering tests, the command, and the output. Then run exactly one scoped re-review of the fix wave (bash "$SKILL_DIR/scripts/review-package" PLAN_FILE FIX_BASE HEAD over the fix range — exit 3 means the fix wave was never committed — and re-review-prompt.md) — for this wave the reviewer reads the plan/spec in place of a task brief and the final fixer's report. Adjudicate any residual findings as in the task loop's breaker: park with rulings, or rule on the load-bearing ones and ledger what you decided. Only the four classes above stop you here. There is no second fix wave — residual load-bearing findings surface to your human partner when sdd-finishing-a-development-branch presents the options.

Finish

Before you delete anything, collect every ledger line containing Ruling: — preflight rulings, parked findings, breaker adjudications, all of them — into your final message under "Rulings I made", in the order you made them, each with what it costs if wrong. Collect every minor (deferred) line into a "Deferred minors" list beside it. Both lists are exhaustive: if the ledger holds a ruling, the list holds it; if it holds a deferred minor, the list holds it. That message is the only place the decisions you took on your human partner's behalf — and the findings you chose not to act on — reach them: they read it and rework whatever you got wrong, or accept what you deferred. A ruling that dies with the workspace was a decision made in secret, and a minor that dies there is a finding silently discarded.

When the final whole-branch review is clean and its fixes are committed, delete this plan's workspace (rm -rf <workspace>) — the git history is the record now. Sibling directories belong to other plans; leave them alone.

Use sdd-finishing-a-development-branch.

Common Rationalizations

Excuse	Reality
"Close enough on spec compliance"	Reviewer found spec gaps = not done. Fix or hit the cap and adjudicate — those are the only exits.
"I'll fix it myself, dispatching is overhead"	Controller fixes pollute your context and skip review. Resume the implementer.
"One more round will converge"	Past the cap, rounds don't converge — the failure is structural. Adjudicate and route.
"The reviewer will just find something new anyway"	Scoped re-reviews verify fixes; they cannot wander. New findings on untouched code go to the ledger, not the loop.
"This finding is obviously wrong, I'll drop it"	You adjudicate only at the cap, and every ruling is a ledger entry. Silent discards are forbidden.
"The fix was small, skip the re-review"	Unreviewed fixes are how regressions land. Every round ends with a scoped re-review.
"Reviews slow the loop down"	The loop without reviews is just unverified churn. Reviews are the loop's brakes and steering.
"Ledger bookkeeping is overhead"	The ledger is what survives compaction. Controllers without one have re-dispatched entire completed task sequences.
"The implementer spawned its own reviewer — free extra assurance"	It's a duplicate seat reviewing the same diff; the task review is the gate. A worker-spawned reviewer is a defect to flag, not rigor.

Example Workflow

You: I'm using Subagent-Driven Development to execute this plan.

[Setup: worktree verified; sdd-workspace resolved; plan + spec read;
 pre-flight table written to ledger; a todo per task]

Task 1: Hook installation script
[task-brief → dispatch implementer with brief + report paths + context]
Implementer asks one question → answered → DONE, tests passing, committed.
[review-package BASE HEAD → task reviewer] Spec ✅, quality approved.
[Ledger: Task 1: complete (commits a1b2c3d..d4e5f6a, review clean)]

Task 2: Recovery modes
[dispatch → DONE, 8/8 passing; review-package → task reviewer]
Reviewer: Spec ❌ — progress reporting missing; Important — magic number (100).
[Fix round 1/5: resume the implementer with both findings; it fixes, re-runs
 the covering tests, and appends its fix report]
[review-package FIX_BASE HEAD → scoped re-review] Both ADDRESSED, no new
 breakage.
[Ledger: Task 2: fix round 1/5 (2 addressed, 0 open; commits d4e5f6a..b7c8d9e)]
[Ledger: Task 2: complete (commits d4e5f6a..b7c8d9e, review clean)]

...
[After all tasks: spec acceptance — each Success Criterion's command or
 recorded procedure run and ledgered; then review-package MERGE_BASE HEAD →
 final reviewer (`sdd-final-reviewer`); ONE fix dispatch + one scoped
 re-review if it finds anything; residuals adjudicated and ledgered]
[Final review clean → delete this plan's workspace]
Done! Using sdd-finishing-a-development-branch.

sdd-subagent-driven-development

SKILL.mdskills/sdd-subagent-driven-development/SKILL.md