Agent skill

Asw Plan

by wjgoarxiv in wjgoarxiv/antigravity-swarm

Antigravity Swarm Plan creates a decision-complete plan before large or ambiguous work.

MITAuto-check passed

Install Asw Plan

skills CLI
$ npx skills add wjgoarxiv/antigravity-swarm --skill asw-plan -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install wjgoarxiv/antigravity-swarm asw-plan --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/wjgoarxiv/antigravity-swarm.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/antigravity-swarm/skills/asw-plan .claude/skills/asw-plan && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
asw-plan
GitHub stars
166
Token cost
~1.4k tokens
SKILL.md length
649 words
Files
1
Skills in repo
15
Repo updated
First seen
Licence
MIT

At a glance

Antigravity Swarm Plan creates a decision-complete plan before large or ambiguous work.

  • Works in 7 steps: Explore the repo before asking questions. → Separate discoverable facts from user… → Decide naming, file ownership, tests, QA… → …
  • SKILL.md covers Planner Duties, Discovery Contract, Required Plan Fields and Downstream Executor Contract, plus 8 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Asw Plan is an agent skill from wjgoarxiv/antigravity-swarm. Antigravity Swarm Plan creates a decision-complete plan before large or ambiguous work.

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: ::A team of AI agents to code for you.::. The licence is MIT.

Example prompts

  • “/asw-plan”

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Explore the repo before asking questions.
  2. Separate discoverable facts from user preferences.
  3. Decide naming, file ownership, tests, QA channels, and rollback boundaries.
  4. Split independent work into Antigravity subagent lanes.
  5. Produce a plan that an executor can follow without inventing missing decisions.
  6. Preserve user language that changes the outcome, but remove rant/noise from the actual task list.
  7. Name every assumption that would be expensive to unwind later.

What it can do on your machine

Read from SKILL.md and the folder at commit a949cb8. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Asw Plan loads about 1.4k tokens when it runs. Until then it costs about 24 tokens; SKILL.md has 649 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~24
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from wjgoarxiv/antigravity-swarm at commit a949cb8, republished under its MIT licence (© wjgoarxiv). 649 words, ~1,360 tokens.

Download SKILL.mdSave it as .claude/skills/asw-plan/SKILL.md (or your agent's skills folder).
name
asw-plan
description
Antigravity Swarm Plan creates a decision-complete plan before large or ambiguous work.

Antigravity Swarm Plan

Use this skill for 5+ step work, migrations, release preparation, package ports, broad refactors, or ambiguous user goals.

You are the planner, not an implementer. Refuse implementation while this skill is active; write the executable plan and direct the caller to start-work.

Planner Duties

  1. Explore the repo before asking questions.
  2. Separate discoverable facts from user preferences.
  3. Decide naming, file ownership, tests, QA channels, and rollback boundaries.
  4. Split independent work into Antigravity subagent lanes.
  5. Produce a plan that an executor can follow without inventing missing decisions.
  6. Preserve user language that changes the outcome, but remove rant/noise from the actual task list.
  7. Name every assumption that would be expensive to unwind later.

Discovery Contract

  • Run parallel repo exploration before drafting; do not plan from memory alone.
  • Use a Socratic interview only for preferences or constraints that cannot be discovered from files.
  • Perform a gap analysis against existing behavior, tests, package surface, and docs before writing tasks.
  • Keep the output to one plan file and do not edit product files in plan mode.
  • Prefer exact local paths over module guesses.
  • If a package, installer, hook, or config surface is involved, include a package-surface audit.

Required Plan Fields

  • Write the plan to .asw/plans/<slug>.md.
  • Include ## TL;DR, objective, non-goals, and decision summary.
  • Include files to edit and tests to write first.
  • Include ## TODOs with atomic checkboxes an executor can run in order.
  • Include QA scenarios with exact commands, expected evidence, and cleanup receipts.
  • Include privacy/package safeguards.
  • End with this exact next step line: Next: start-work <plan-name>.

Downstream Executor Contract

Each checkbox must include references, acceptance criteria, test-first instructions, real-surface QA, expected evidence paths, cleanup receipts, and commit guidance. The executor should be able to resume from the plan without interviewing the user again.

Never end with a passive handoff. End with the exact next step instead.

Parallel Execution Waves

Split work into waves when tasks can proceed independently.

text
Wave 1:
- Task A: no dependencies
- Task B: no dependencies

Wave 2:
- Task C: depends on A

Critical path:
A -> C

A good wave plan:

  • keeps shared setup early,
  • avoids two agents editing the same file,
  • groups independent tests and docs separately,
  • reserves final verification for after all implementation waves,
  • makes blockers explicit.

If fewer than two tasks can run independently, say so and keep execution serialized.

Dependency Matrix

Every non-trivial plan needs a matrix:

TaskDepends onBlocksCan parallelize with
1none32

The executor should not need to infer ordering from prose.

Show full SKILL.md (252 more words)Show less

Task Contract

Every task must include:

  • what to do,
  • what not to do,
  • files or directories,
  • references,
  • RED test,
  • GREEN check,
  • manual QA scenario,
  • cleanup receipt,
  • acceptance criteria,
  • commit guidance.

Do not separate "write code" and "write tests" into different tasks. Implementation and verification belong together.

QA Scenarios

Each task needs at least one concrete scenario:

text
Scenario:
Channel:
Steps:
Expected:
Evidence:
Cleanup:

Use:

  • HTTP call for API behavior,
  • tmux for CLI/install/HUD/hook behavior,
  • browser for web behavior,
  • computer use for desktop or IDE behavior.

The scenario must be agent-executable. Do not write "user manually checks".

Commit Strategy

Give commit guidance per task:

text
Commit: YES|NO
Message:
Files:
Reason:

Use conventional commit subjects. Keep commits atomic and green. If the user did not ask for commits, mark commit as NO and instruct the executor to report a draft message instead.

Final Verification Wave

The plan must end with a verification wave:

  • plan compliance audit,
  • full automated tests,
  • package/private surface scan,
  • plugin validation when applicable,
  • manual QA replay for user-visible surfaces,
  • reviewer pass when the change is broad or risky,
  • git status review,
  • cleanup receipt review.

This wave is not optional. It is the guard against "green tests, broken product".

Plan Shape

Use this skeleton unless the repository clearly needs something else:

markdown
# <plan-name>

## TL;DR
<one paragraph>

## Objective
<observable outcome>

## Non-goals
- <what must not change>

## Discovery
- <path>: <fact>

## Decisions
- <decision>: <reason>

## TODOs
- [ ] <task>
  - Files:
  - RED:
  - GREEN:
  - Real-surface QA:
  - Evidence:
  - Cleanup:
  - Commit:

## Parallel Execution Waves
- Wave 1:

## Dependency Matrix
| Task | Depends on | Blocks | Can parallelize with |
|---|---|---|---|

## Final Verification Wave
- [ ] <full test/package/docs/manual QA pass>

Next: `start-work <plan-name>`

Quality Bar

  • A plan that says only "create implementation, then test" is not executable.
  • A plan that lacks a real-surface check is incomplete when users can observe the change.
  • A plan that writes broad tasks without file ownership will stall the executor.
  • A plan that requires hidden context from the planning chat is not acceptable.

© wjgoarxiv, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in plugins/antigravity-swarm/skills/asw-plan of wjgoarxiv/antigravity-swarm.

Open the folder on GitHubat commit a949cb8

Compare with similar skills

Asw Plan next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Asw Plan compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Asw Plan this skillwjgoarxiv/antigravity-swarm166—~1.4kAutomated safety check: PassMIT
Agent Swarmruvnet/ruflo74k2 repos~891Automated safety check: PassMIT
Swarm Initruvnet/ruflo74k—~319Automated safety check: PassMIT
Agent Release Swarmruvnet/ruflo74k2 repos~3.5kAutomated safety check: PassMIT
Agent Swarm Issueruvnet/ruflo74k2 repos~3.6kAutomated safety check: PassMIT
Swarm Orchestrationruvnet/ruflo74k—~261Automated safety check: PassMIT

Similar skills

  • Agent Swarm

    ruvnet/ruflo

    Agent skill for swarm - invoke with $agent-swarm. An agent skill from ruvnet/ruflo.

    74k GitHub starsUsed in 2 repos~891 tokens
    Agent WorkflowsAuto-check passed
  • Swarm Init

    ruvnet/ruflo

    Initialize a multi-agent swarm with anti-drift configuration.

    74k GitHub stars~319 tokensUpdated yesterday
    SecurityAuto-check passed
  • Agent skill for release-swarm - invoke with $agent-release-swarm

    74k GitHub starsUsed in 2 repos~3.5k tokens
    DevelopmentAuto-check passed
  • Agent Swarm Issue

    ruvnet/ruflo

    Agent skill for swarm-issue - invoke with $agent-swarm-issue

    74k GitHub starsUsed in 2 repos~3.6k tokens
    Agent WorkflowsAuto-check passed
  • Multi-agent swarm coordination for complex tasks. An agent skill from ruvnet/ruflo.

    74k GitHub stars~261 tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Coordinates a hierarchical swarm of specialized agents through the claude-flow CLI for work that spans several files or modules at once.

    74k GitHub starsUsed in 1 repo~779 tokens
    Agent WorkflowsAuto-check passed

More from wjgoarxiv/antigravity-swarm

All 15 skills in this repo
  • Asw

    wjgoarxiv/antigravity-swarm

    Antigravity Swarm execution loop for evidence-driven implementation with tests, subagents, hooks, and manual QA.

    166 GitHub stars~1.1k tokensUpdated 3 mo ago
    Auto-check passed
  • Asw Cleanup

    wjgoarxiv/antigravity-swarm

    Remove AI-looking clutter and temporary artifacts without changing behavior.

    166 GitHub stars~1.2k tokensUpdated 3 mo ago
    Auto-check passed
  • Asw Debug

    wjgoarxiv/antigravity-swarm

    Hypothesis-driven Antigravity Swarm debugging for crashes, hangs, wrong output, and runtime drift.

    166 GitHub stars~1k tokensUpdated 3 mo ago
    Auto-check passed
  • Asw Goal

    wjgoarxiv/antigravity-swarm

    Durable Antigravity Swarm goal orchestration with explicit success criteria, evidence ledger, manual QA channels, and completion audit.

    166 GitHub stars~905 tokensUpdated 3 mo ago
    Auto-check passed
  • Asw Loop

    wjgoarxiv/antigravity-swarm

    Antigravity Swarm Loop executes RED to GREEN to real-surface QA with cleanup receipts.

    166 GitHub stars~824 tokensUpdated 3 mo ago
    Auto-check passed
  • Asw Refactor

    wjgoarxiv/antigravity-swarm

    Behavior-preserving Antigravity Swarm refactoring with characterization, impact mapping, diagnostics, and real-surface verification.

    166 GitHub stars~3.9k tokensUpdated 3 mo ago
    Auto-check passed

Questions about Asw Plan

What does Asw Plan do?

Antigravity Swarm Plan creates a decision-complete plan before large or ambiguous work. Asw Plan is an agent skill from wjgoarxiv/antigravity-swarm. Antigravity Swarm Plan creates a decision-complete plan before large or ambiguous work.

How do I install Asw Plan in Claude Code?

Run `npx skills add wjgoarxiv/antigravity-swarm --skill asw-plan -a claude-code`. Or copy the skill folder (plugins/antigravity-swarm/skills/asw-plan in wjgoarxiv/antigravity-swarm) into .claude/skills/asw-plan in your project. Claude Code loads it when a task matches its description.

How do I install Asw Plan in Codex?

Run `npx skills add wjgoarxiv/antigravity-swarm --skill asw-plan -a codex`. Or copy the skill folder (plugins/antigravity-swarm/skills/asw-plan in wjgoarxiv/antigravity-swarm) into .agents/skills/asw-plan in your project. Codex loads it when a task matches its description.

Can I use Asw Plan in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add wjgoarxiv/antigravity-swarm --skill asw-plan -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/asw-plan, .gemini/skills/asw-plan, .github/skills/asw-plan and .opencode/skills/asw-plan in your project.

What does Asw Plan need to run?

SKILL.md names no scripts, command-line tools or credentials: Asw Plan is instructions for the agent only.

Does Asw Plan access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Asw Plan safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Asw Plan use?

Asw Plan is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Asw Plan use?

About 1.4k tokens (SKILL.md is roughly 5.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Asw Plan?

Skills that share tags, products or a category with Asw Plan: Agent Swarm (ruvnet/ruflo, 74k stars), Swarm Init (ruvnet/ruflo, 74k stars), Agent Release Swarm (ruvnet/ruflo, 74k stars) and Agent Swarm Issue (ruvnet/ruflo, 74k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Asw Plan?

wjgoarxiv (a GitHub user) maintains it in wjgoarxiv/antigravity-swarm, which has 166 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on June 19, 2026.

Source: wjgoarxiv/antigravity-swarm on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.