Skip to content

feat(skills): add Batch Execution, SDD, and Writing Plans skills#19

Open
maystudios wants to merge 1 commit intomainfrom
worktree-agent-ae748c3c
Open

feat(skills): add Batch Execution, SDD, and Writing Plans skills#19
maystudios wants to merge 1 commit intomainfrom
worktree-agent-ae748c3c

Conversation

@maystudios
Copy link
Owner

Summary

  • Add Batch Execution skill: parallel agent spawning in isolated git worktrees for 5-30 independent work units
  • Add Subagent-Driven Development (SDD) skill: fresh subagent per task with 2-stage review to prevent context rot
  • Add Writing Plans skill: standardized PLAN.md format with verify/done blocks, dependency detection, and wave grouping

Test plan

  • Unit tests pass (npx vitest run)
  • Full build succeeds (npm run build)
  • New skills appear in dist/assets/templates/skills/ after build (140 template files copied)
  • Verify skill-context batch-execution, skill-context subagent-driven-development, and skill-context writing-plans CLI commands load the new skills

🤖 Generated with Claude Code

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Copilot AI review requested due to automatic review settings March 2, 2026 04:51
Copy link

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Adds three new MAXSIM “skill” templates intended to standardize planning and execution workflows (parallel batch execution, subagent-per-task execution, and PLAN.md authoring).

Changes:

  • Add batch-execution skill template describing parallel work via isolated git worktrees.
  • Add subagent-driven-development skill template describing one-task-per-subagent with inter-task review.
  • Add writing-plans skill template describing a standardized PLAN.md authoring format with verify/done gates, dependency detection, and wave grouping.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 7 comments.

File Description
templates/skills/batch-execution/SKILL.md New skill template for parallelizing independent units of work.
templates/skills/subagent-driven-development/SKILL.md New skill template for executing each plan task with a fresh subagent and 2-stage review.
templates/skills/writing-plans/SKILL.md New skill template for writing structured plans with explicit verification and done criteria.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

### Context Loading

```bash
node ~/.claude/maxsim/bin/maxsim-tools.cjs skill-context subagent-driven-development
Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This references a skill-context subcommand on maxsim-tools, but the CLI command registry (packages/cli/src/cli.ts:305-370) doesn’t include skill-context. Unless this command exists elsewhere, this example won’t work; either implement the command or adjust the docs to point to the installed skill location (e.g., under ~/.claude/maxsim/agents/skills/<skill>/SKILL.md).

Suggested change
node ~/.claude/maxsim/bin/maxsim-tools.cjs skill-context subagent-driven-development
# View this skill's installed documentation:
cat ~/.claude/maxsim/agents/skills/subagent-driven-development/SKILL.md

Copilot uses AI. Check for mistakes.
Comment on lines +70 to +77
| Excuse | Why It Violates the Rule |
|--------|--------------------------|
| "Sequential is simpler" | Simpler for you, slower for the user. Parallel is the job. |
| "What if they conflict?" | Verify independence first (step 2). If they conflict, they are not independent — re-decompose. |
| "Too many agents" | The threshold is 5 units. Below 5, sequential is acceptable. Above 5, parallelize. |
| "I'll batch the small ones" | Small units are the easiest to parallelize. Do not combine them. |
| "Worktrees are complex" | Worktree creation is automated. Your job is decomposition and monitoring. |

Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The “Common Rationalizations” table is using || at the start of each row, which renders as an extra empty column in standard Markdown. Other skills use normal 2‑column tables (e.g., templates/skills/tdd/SKILL.md:68-75). Consider switching these rows to a single leading | so the table renders consistently.

Copilot uses AI. Check for mistakes.
### Context Loading

```bash
node ~/.claude/maxsim/bin/maxsim-tools.cjs skill-context batch-execution
Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This references a skill-context subcommand on maxsim-tools, but the CLI command registry (packages/cli/src/cli.ts:305-370) doesn’t include skill-context. Unless this command exists elsewhere, this example won’t work; either implement the command or adjust the docs to point to the installed skill location (e.g., under ~/.claude/maxsim/agents/skills/<skill>/SKILL.md).

Suggested change
node ~/.claude/maxsim/bin/maxsim-tools.cjs skill-context batch-execution
# View the installed batch-execution skill context
cat ~/.claude/maxsim/agents/skills/batch-execution/SKILL.md

Copilot uses AI. Check for mistakes.
Comment on lines +125 to +132
| Excuse | Why It Violates the Rule |
|--------|--------------------------|
| "The executor is smart enough to figure it out" | Smart executors with vague plans produce inconsistent results. Be explicit. |
| "Verify blocks are obvious" | If they are obvious, writing them takes 10 seconds. Do it. |
| "Done criteria are implicit in the description" | Implicit criteria cannot be checked. Make them explicit checkboxes. |
| "Dependencies are clear from context" | Context dies between agents. Write dependencies explicitly. |
| "Wave grouping is premature optimization" | Wave grouping enables parallel execution. It is not optimization — it is correctness. |
| "The plan is too detailed" | Plans cannot be too detailed. They can only be too vague. |
Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The “Common Rationalizations” table is using || at the start of each row, which renders as an extra empty column in standard Markdown. Other skills use normal 2‑column tables (e.g., templates/skills/tdd/SKILL.md:68-75). Consider switching these rows to a single leading | so the table renders consistently.

Copilot uses AI. Check for mistakes.
Comment on lines +94 to +121
## Plan Structure Template

```markdown
# Phase [N] Plan [M]: [Phase Name]

## Overview
[1-2 sentences: what this plan achieves and why]

## Tasks

### Task 1: [Title]
...

### Task 2: [Title]
...

## Execution Order

**Wave 1** (parallel): Tasks 1, 2
**Wave 2** (parallel): Tasks 3, 4
**Wave 3** (sequential): Task 5

## Verification

After all tasks complete:
- [ ] [Phase-level success criterion 1]
- [ ] [Phase-level success criterion 2]
```
Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The “Plan Structure Template” here doesn’t include the XML-like <task>…</task> blocks that MAXSIM tooling parses/validates (see packages/cli/src/core/template.ts:176-223 and packages/cli/src/core/verify.ts:284-305). As written, maxsim-tools verify plan-structure will warn “No elements found” and the dashboard task parser won’t see any tasks. Please update this template to match the actual plan format (YAML frontmatter + <task> with <name>, <files>, <action>, <verify>, <done>).

Copilot uses AI. Check for mistakes.
### Context Loading

```bash
node ~/.claude/maxsim/bin/maxsim-tools.cjs skill-context writing-plans
Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This references a skill-context subcommand on maxsim-tools, but the CLI command registry (packages/cli/src/cli.ts:305-370) doesn’t include skill-context. Unless this command exists elsewhere, this example won’t work; either implement the command or adjust the docs to point to the installed skill location (e.g., under ~/.claude/maxsim/agents/skills/<skill>/SKILL.md).

Suggested change
node ~/.claude/maxsim/bin/maxsim-tools.cjs skill-context writing-plans
# View the installed skill documentation for writing-plans
cat ~/.claude/maxsim/agents/skills/writing-plans/SKILL.md

Copilot uses AI. Check for mistakes.
Comment on lines +68 to +75
| Excuse | Why It Violates the Rule |
|--------|--------------------------|
| "I have context from the last task" | That context is rotting. Fresh agent, fresh perspective. |
| "Spawning agents is overhead" | Context rot costs more than spawn time. The math is clear. |
| "I can do multiple tasks efficiently" | Efficiency without accuracy is waste. One task, one agent. |
| "The tasks are related" | Related tasks still get separate agents. Share via committed code, not session state. |
| "Review between tasks slows things down" | Review catches what rot misses. The slowdown is an investment. |

Copy link

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The “Common Rationalizations” table is using || at the start of each row, which renders as an extra empty column in standard Markdown. Other skills use normal 2‑column tables (e.g., templates/skills/tdd/SKILL.md:68-75). Consider switching these rows to a single leading | so the table renders consistently.

Copilot uses AI. Check for mistakes.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants