feat(skills/developing-a-feature): conditional task-level SDD on the single-PR path#13
Merged
Merged
Conversation
…single-PR path The single-PR path implemented directly with no internal review until the external PR reviewer. When the committed plan has 2+ independent tasks, Step 4 now drives the build through superpowers:subagent-driven-development (fresh implementer subagent per task, then orchestrator-run review, fix-loop, continuous execution) — the in-session analogue of the multi-PR fan-out's author-and-reviewer-are-different-contexts discipline. Per-task review uses the local `review` skill (spec-compliance gate, then code quality), consistent with fanning-out-with-worktrees. Below the threshold — one cohesive change or a strictly sequential chain — the direct TDD path is kept, because SDD's dispatch overhead only pays off on independent tasks. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
Pull request overview
Updates the developing-a-feature skill’s Step 4 guidance to add an explicit decision point for Single-PR implementation: when a committed plan contains multiple independent tasks, it routes execution through a task-by-task subagent loop with orchestrator-driven review to avoid “same-context author+reviewer” gaps.
Changes:
- Makes the Single-PR Step 4 path conditional: 2+ independent tasks (excluding the integration task) uses
superpowers:subagent-driven-development; otherwise keep the direct orchestrator TDD path. - Specifies per-task review mechanics using the
reviewskill in two passes (spec gate, then code quality), routed back to the implementer subagent. - Adds new “Red flags” rows addressing common failure modes for independent-task plans in a single PR.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| - Run the project's test and lint commands (and typecheck, if it has one) before claiming work is done. Discover them from the project's CLAUDE.md / AGENTS.md or its build config (Makefile, package.json, etc.). | ||
| - **Single-PR** — the build mode depends on the plan's task shape. Read the committed plan's task list and count the tasks that are **independent** (no task depends on another's output — they could, in principle, be built in any order). The integration task that ties them together does not count toward independence; it consumes the others. | ||
|
|
||
| - **2+ independent tasks → drive the build task-by-task with review between tasks.** **REQUIRED SUB-SKILL:** `superpowers:subagent-driven-development`. The orchestrator (this session — the main loop, which *can* dispatch subagents) runs that skill's loop engine: a fresh implementer subagent per task with full task text handed in, then review, fix-loop, and continuous execution across all tasks. Two adaptations keep it consistent with the rest of this plugin: |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
The single-PR path implemented directly with no internal review until the external PR reviewer — the emptiest review spot in the workflow. When the committed plan has 2+ independent tasks, Step 4 now drives the build through
superpowers:subagent-driven-development(fresh implementer subagent per task, orchestrator-run review, fix-loop, continuous execution) — the in-session analogue of the multi-PR fan-out's "author and reviewer are different contexts" discipline.Changes
developing-a-featureStep 4 Single-PR branch is now conditional: 2+ independent tasks (the integration task excluded) → delegate tosuperpowers:subagent-driven-development; one cohesive change or a strictly sequential chain → the existing direct TDD path.reviewskill (spec-compliance gate, then code quality), consistent withfanning-out-with-worktrees.Testing
Built via
superpowers:writing-skillswith isolated baselines. RED baselines defaulted to implementing all tasks themselves with no per-task review, deferring review to the end or the human, and let "do it myself" override the task-independence signal. With the skill, an isolated orchestrator chose SDD for a 4-task independent plan (counting 3 independent, excluding the integration task) and direct TDD for a 3-task sequential chain — the conditional discriminates correctly.Related
Shares a reviewer-mechanism choice (local
reviewskill) with the two-stage sub-PR review change (#13); touchesdeveloping-a-featurealongside the held review-loop gates change, in a different section.