New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Parallel Explore

by YougLin-dev in YougLin-dev/Aha-Loop

Guides parallel exploration of multiple implementation approaches using git worktrees.

MITAuto-check passedBackend & APIs

Install Parallel Explore

skills CLI
$ npx skills add YougLin-dev/Aha-Loop --skill parallel-explore -a claude-code

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

GitHub CLI
$ gh skill install YougLin-dev/Aha-Loop parallel-explore --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/YougLin-dev/Aha-Loop.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/parallel-explore .claude/skills/parallel-explore && 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
parallel-explore
GitHub stars
181
Token cost
~1.9k tokens
SKILL.md length
533 words
Files
1
Skills in repo
15
Repo updated
First seen
Licence
MIT

At a glance

Guides parallel exploration of multiple implementation approaches using git worktrees.

  • Works in 7 steps: Identify the Decision Point → Start Exploration → Work in Each Worktree → …
  • Facing decisions with multiple valid paths
  • SKILL.md covers The Job, When to Use Parallel Exploration, Exploration Process and Parallel Exploration Mindset, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Parallel Explore is an agent skill from YougLin-dev/Aha-Loop. Guides parallel exploration of multiple implementation approaches using git worktrees. Use when facing decisions with multiple valid paths. Triggers on: explore options, compare approaches, parallel exploration.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Backend & APIs, covering Git worktrees and Authentication. The repository describes itself as: [MVP] Aha Loop is a fully autonomous AI development system, extended from the core ideas of Ralph. It's not just an execution engine, but a complete AI development framework with… The licence is MIT.

When your agent uses it

  • Facing decisions with multiple valid paths
  • : explore options
  • Compare approaches
  • Parallel exploration

Example prompts

  • “Use the parallel-explore skill to guide parallel exploration of multiple implementation approaches using git worktrees”
  • “/parallel-explore”

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. Identify the Decision Point
  2. Start Exploration
  3. Work in Each Worktree
  4. Create Exploration Result
  5. Evaluate Results
  6. Merge Best Approach
  7. Cleanup

What it can do on your machine

Read from SKILL.md and the folder at commit 8d799b2. 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 bash and 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

Parallel Explore loads about 1.9k tokens when it runs. Until then it costs about 57 tokens; SKILL.md has 533 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~57
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k

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 YougLin-dev/Aha-Loop at commit 8d799b2, republished under its MIT licence (© YougLin-dev). 533 words, ~1,932 tokens.

Download SKILL.mdSave it as .claude/skills/parallel-explore/SKILL.md (or your agent's skills folder).
name
parallel-explore
description
Guides parallel exploration of multiple implementation approaches using git worktrees. Use when facing decisions with multiple valid paths. Triggers on: explore options, compare approaches, parallel exploration.

Parallel Exploration Skill

Guide the process of exploring multiple implementation approaches simultaneously using git worktrees.

The Job

  1. Identify when parallel exploration would be valuable
  2. Define distinct approaches to explore
  3. Create isolated worktrees for each approach
  4. Execute exploration in parallel
  5. Evaluate and compare results
  6. Merge the best solution

When to Use Parallel Exploration

Good Candidates
  • Architecture decisions - Different patterns (e.g., microservices vs monolith)
  • Library selection - Comparing similar libraries hands-on
  • Algorithm choices - Different approaches to the same problem
  • API design - Different interface designs
  • Performance optimization - Multiple optimization strategies
Not Worth Parallelizing
  • Simple, clear-cut decisions
  • Tasks with obvious single approach
  • Very small changes
  • Changes that don't warrant the overhead

Exploration Process

Step 1: Identify the Decision Point

When you encounter a significant decision:

markdown
## Decision Point Identified

**Question:** [What needs to be decided]
**Context:** [Why this matters]
**Approaches to Explore:**
1. [Approach A] - [Brief description]
2. [Approach B] - [Brief description]
3. [Approach C] - [Brief description]

**Exploration Value:** [Why parallel exploration helps here]
Step 2: Start Exploration

Use the parallel explorer script:

bash
./scripts/aha-loop/parallel-explorer.sh explore "task description" --approaches "approach1,approach2,approach3"

Or let AI suggest approaches:

bash
./scripts/aha-loop/parallel-explorer.sh explore "task description"
# AI will suggest approaches automatically
Step 3: Work in Each Worktree

In each worktree, the AI should:

  1. Implement fully - Not just a stub, but working code
  2. Write tests - Validate the approach works
  3. Document findings - Create EXPLORATION_RESULT.md
Step 4: Create Exploration Result

Each worktree must have EXPLORATION_RESULT.md:

markdown
# Exploration Result: [Approach Name]

## Summary
[What was implemented]

## Implementation Details
- [Key implementation decisions]
- [Code structure choices]
- [Dependencies used]

## Pros Discovered
- [Pro 1]
- [Pro 2]
- ...

## Cons Discovered
- [Con 1]
- [Con 2]
- ...

## Code Quality Assessment
- Maintainability: [1-10]
- Readability: [1-10]
- Testability: [1-10]
- Performance: [1-10]

## Unexpected Findings
[Anything surprising learned during implementation]

## Recommendation Score
[1-10] - [Brief justification]

## Would Need If Chosen
[What additional work would be needed to production-ready this approach]
Step 5: Evaluate Results

After all explorations complete:

bash
./scripts/aha-loop/parallel-explorer.sh evaluate explore-task-12345

This triggers multiple evaluation agents who:

  • Review all EXPLORATION_RESULT.md files
  • Compare approaches objectively
  • Discuss trade-offs
  • Make final recommendation
Step 6: Merge Best Approach
bash
./scripts/aha-loop/parallel-explorer.sh merge explore-task-12345 chosen-approach
Step 7: Cleanup
bash
./scripts/aha-loop/parallel-explorer.sh cleanup explore-task-12345

Parallel Exploration Mindset

Unlimited Resources

You have unlimited computational resources. Don't hold back:

  • Explore as many approaches as seem valuable
  • Run multiple variations of the same approach
  • Don't worry about "wasting" resources on exploration
Genuine Exploration

Each approach should be explored genuinely:

  • Don't sabotage approaches you don't prefer
  • Give each approach fair effort
  • Document honestly, including when an approach works better than expected
Learn from All Approaches

Even "losing" approaches provide value:

  • Document why they didn't work
  • Note any good ideas to borrow
  • Record lessons for future decisions

Commands Reference

bash
# Start exploration
./scripts/aha-loop/parallel-explorer.sh explore "description" --approaches "a,b,c"

# Check status
./scripts/aha-loop/parallel-explorer.sh status
./scripts/aha-loop/parallel-explorer.sh status explore-task-12345

# Evaluate completed explorations
./scripts/aha-loop/parallel-explorer.sh evaluate explore-task-12345

# List all worktrees
./scripts/aha-loop/parallel-explorer.sh list

# Merge chosen approach
./scripts/aha-loop/parallel-explorer.sh merge explore-task-12345 chosen-approach

# Cleanup
./scripts/aha-loop/parallel-explorer.sh cleanup explore-task-12345
./scripts/aha-loop/parallel-explorer.sh cleanup --all

Example: Exploring Authentication Strategies

Trigger

You're implementing user authentication and there are multiple valid approaches.

Show full SKILL.md (210 more words)Show less
Start Exploration
bash
./scripts/aha-loop/parallel-explorer.sh explore "user authentication for web app" \
  --approaches "jwt-stateless,session-based,magic-link"
In JWT Worktree

Implement full JWT authentication:

  • Token generation
  • Token validation
  • Refresh token flow
  • Tests

Create EXPLORATION_RESULT.md documenting:

  • Pros: Stateless, scalable, mobile-friendly
  • Cons: Token invalidation complexity, larger requests
  • Score: 7/10
In Session Worktree

Implement session-based auth:

  • Session creation
  • Session storage (Redis/DB)
  • Session middleware
  • Tests

Create EXPLORATION_RESULT.md documenting:

  • Pros: Simple to invalidate, smaller requests
  • Cons: Requires session store, stateful
  • Score: 8/10

Implement passwordless magic link:

  • Email sending
  • Link generation
  • Link validation
  • Tests

Create EXPLORATION_RESULT.md documenting:

  • Pros: No password to manage, good UX
  • Cons: Email dependency, potential delays
  • Score: 6/10
Evaluate

Multiple evaluation agents review all three approaches and produce:

  • Comparison table
  • Final recommendation (session-based for simplicity)
  • Suggestion to borrow JWT's token rotation idea
Merge
bash
./scripts/aha-loop/parallel-explorer.sh merge explore-auth-12345 session-based

Nested Exploration

During exploration, you may discover new decision points. You can:

  1. Defer - Note it for later, continue with current exploration
  2. Nest - Start a sub-exploration within the worktree

For nested exploration, create a sub-directory:

.worktrees/
  explore-auth-12345/
    jwt/
      .worktrees/          # Nested exploration
        explore-refresh-67890/
          rotating-tokens/
          sliding-window/

Integration with Observability

Log exploration decisions to logs/ai-thoughts.md:

markdown
## 2026-01-29 14:00:00 | Task: PRD-003 | Phase: Exploration

### Decision Point
Authentication strategy needs to be determined.

### Approaches Being Explored
1. JWT (stateless)
2. Session-based (stateful)
3. Magic link (passwordless)

### Exploration Status
Started parallel exploration: explore-auth-12345

### Expected Outcome
Will evaluate all three and merge best approach.

Checklist

Before starting exploration:

  • Decision point clearly identified
  • Multiple valid approaches exist
  • Approaches are meaningfully different
  • Exploration effort is justified

During exploration:

  • Each approach implemented genuinely
  • Tests written for each
  • EXPLORATION_RESULT.md created

After exploration:

  • All approaches evaluated fairly
  • Final recommendation documented
  • Best approach merged
  • Worktrees cleaned up

© YougLin-dev, 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 .agents/skills/parallel-explore of YougLin-dev/Aha-Loop.

Open the folder on GitHubat commit 8d799b2

Compare with similar skills

Parallel Explore 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.

Parallel Explore compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Parallel Explore this skillYougLin-dev/Aha-Loop181—~1.9kAutomated safety check: PassMIT
Project Session ManagementMicrock/ordinary-claude-skills401—~1.9kAutomated safety check: PassCustom licence
Teams Devmicrosoft/teams-sdk728—~1.2kAutomated safety check: PassMIT
React AdminAwesome-Technologies/synapse-admin1.2k—~2.6kAutomated safety check: PassApache-2.0
Tabz TerminalsGGPrompts/TabzChrome147—~696Automated safety check: PassMIT
API Debuggingownpilot/OwnPilot426—~824Automated safety check: PassMIT

Similar skills

  • Project Session Management

    Microck/ordinary-claude-skills

    Track progress across work sessions using SESSION.md with git checkpoints and concrete next actions.

    401 GitHub stars~1.9k tokensUpdated 1 mo ago
    Backend & APIsAuto-check passed

  • Teams Dev

    microsoft/teams-sdk

    Official

    A skill your agent uses whenever the user mentions Microsoft Teams in a development context — whether they're building, integrating, configuring, debugging, or just asking questions.

    728 GitHub stars~1.2k tokensUpdated 2 days ago
    Backend & APIsAuto-check passed

  • React Admin

    Awesome-Technologies/synapse-admin

    This skill should be used when building, modifying, or debugging a react-admin application — including creating resources, lists, forms, data fetching, authentication, relationships between…

    1.2k GitHub stars~2.6k tokensUpdated 20 days ago
    Backend & APIsAuto-check passed

  • Tabz Terminals

    GGPrompts/TabzChrome

    Spawn and manage terminal tabs via TabzChrome REST API. An agent skill from GGPrompts/TabzChrome.

    147 GitHub stars~696 tokensUpdated 10 days ago
    Backend & APIsAuto-check passed

  • API Debugging

    ownpilot/OwnPilot

    Systematic approach to debugging REST APIs, HTTP errors, authentication issues, and network problems.

    426 GitHub stars~824 tokensUpdated 29 days ago
    Backend & APIsAuto-check passed

  • Tnr Dev Server

    studie-tech/TheNinjaRPG

    Run a TheNinjaRPG dev server locally from any git worktree, provision disposable test users, and call tRPC endpoints as those users.

    104 GitHub stars~2.5k tokensUpdated yesterday
    DevelopmentAuto-check: notes

More from YougLin-dev/Aha-Loop

All 15 skills in this repo

  • God Member

    YougLin-dev/Aha-Loop

    Defines God Committee member behavior and responsibilities with oversight authority.

    181 GitHub starsUsed in 1 repo~1.6k tokens
    Auto-check passed

  • Architect

    YougLin-dev/Aha-Loop

    Designs system architecture and selects technology stack based on vision analysis.

    181 GitHub stars~2.2k tokensUpdated 8 mo ago
    Auto-check passed

  • Doc Review

    YougLin-dev/Aha-Loop

    Reviews and cleans up outdated documentation. An agent skill from YougLin-dev/Aha-Loop.

    181 GitHub stars~1.5k tokensUpdated 8 mo ago
    Auto-check passed

  • God Consensus

    YougLin-dev/Aha-Loop

    Guides God Committee members through consensus-building for collective decisions.

    181 GitHub stars~2k tokensUpdated 8 mo ago
    Auto-check passed

  • God Intervention

    YougLin-dev/Aha-Loop

    Guides God Committee members through executing interventions.

    181 GitHub stars~2.3k tokensUpdated 8 mo ago
    Auto-check passed

  • Plan Review

    YougLin-dev/Aha-Loop

    Reviews and adjusts PRD plans based on research findings. An agent skill from YougLin-dev/Aha-Loop.

    181 GitHub stars~1.8k tokensUpdated 8 mo ago
    Auto-check passed

Categories

Questions about Parallel Explore

What does Parallel Explore do?

Guides parallel exploration of multiple implementation approaches using git worktrees. Parallel Explore is an agent skill from YougLin-dev/Aha-Loop. Guides parallel exploration of multiple implementation approaches using git worktrees.

When should I use Parallel Explore?

Parallel Explore fits situations like: facing decisions with multiple valid paths; : explore options; compare approaches; parallel exploration.

How do I install Parallel Explore in Claude Code?

Run `npx skills add YougLin-dev/Aha-Loop --skill parallel-explore -a claude-code`. Or copy the skill folder (.agents/skills/parallel-explore in YougLin-dev/Aha-Loop) into .claude/skills/parallel-explore in your project. Claude Code loads it when a task matches its description.

How do I install Parallel Explore in Codex?

Run `npx skills add YougLin-dev/Aha-Loop --skill parallel-explore -a codex`. Or copy the skill folder (.agents/skills/parallel-explore in YougLin-dev/Aha-Loop) into .agents/skills/parallel-explore in your project. Codex loads it when a task matches its description.

Can I use Parallel Explore 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 YougLin-dev/Aha-Loop --skill parallel-explore -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/parallel-explore, .gemini/skills/parallel-explore, .github/skills/parallel-explore and .opencode/skills/parallel-explore in your project.

What does Parallel Explore need to run?

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

Does Parallel Explore 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 Parallel Explore 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 Parallel Explore use?

Parallel Explore 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 Parallel Explore use?

About 1.9k tokens (SKILL.md is roughly 7.7k 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 Parallel Explore?

Skills that share tags, products or a category with Parallel Explore: Project Session Management (Microck/ordinary-claude-skills, 401 stars), Teams Dev (microsoft/teams-sdk, 728 stars), React Admin (Awesome-Technologies/synapse-admin, 1.2k stars) and Tabz Terminals (GGPrompts/TabzChrome, 147 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Parallel Explore?

YougLin-dev (a GitHub user) maintains it in YougLin-dev/Aha-Loop, which has 181 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on February 3, 2026.

Source: YougLin-dev/Aha-Loop on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.