New: 98,000+ 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

Run Parallel Tasks

by LIDR-academy in LIDR-academy/AI4Devs-LTI-extended

Run N feature tasks in parallel, each in its own worktree, following the full specboot pipeline (enrich → new → ff → apply → verify).

MITAuto-check passedProduct & Project Management

Install Run Parallel Tasks

skills CLI
$ npx skills add LIDR-academy/AI4Devs-LTI-extended --skill run-parallel-tasks -a claude-code

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

GitHub CLI
$ gh skill install LIDR-academy/AI4Devs-LTI-extended run-parallel-tasks --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/LIDR-academy/AI4Devs-LTI-extended.git skills-src && mkdir -p .claude/skills && cp -r skills-src/ai-specs/skills/run-parallel-tasks .claude/skills/run-parallel-tasks && 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
run-parallel-tasks
GitHub stars
278
Token cost
~1.5k tokens
SKILL.md length
450 words
Files
2 (incl. references)
Skills in repo
9
Repo updated
First seen
Licence
MIT

At a glance

Run N feature tasks in parallel, each in its own worktree, following the full specboot pipeline (enrich → new → ff → apply → verify).

  • Works in 4 steps: Resolve task source (arguments first,… → Enrich US for each task → Spawn one agent per task in parallel → …
  • Tasks that involve Git worktrees
  • SKILL.md covers When to invoke, Instructions and Agent prompt template
  • Calls git

What it does

Run Parallel Tasks is an agent skill from LIDR-academy/AI4Devs-LTI-extended. Run N feature tasks in parallel, each in its own worktree, following the full specboot pipeline (enrich → new → ff → apply → verify). Stops after verify — no archive, no commit, no cleanup. Explicit task arguments override parallel-tasks.md; file is fallback only.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/parallel-tasks.md`).

It sits in Product & Project Management, covering Git worktrees. It works with Jira. The repository describes itself as: Repository with several experiments from live sessions. The licence is MIT.

When your agent uses it

  • Tasks that involve Git worktrees

Example prompts

  • “/run-parallel-tasks”

Workflow steps

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

  1. Resolve task source (arguments first, file fallback)
  2. Enrich US for each task
  3. Spawn one agent per task in parallel
  4. Wait and report

What it can do on your machine

Read from SKILL.md and the folder at commit 9ff9a80. 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

    Shell commands in SKILL.md call:

    • git

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

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    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

Run Parallel Tasks loads about 1.5k tokens when it runs, and up to ~1.8k if it reads all its reference files. Until then it costs about 71 tokens; SKILL.md has 450 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~71
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~1.8k

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 LIDR-academy/AI4Devs-LTI-extended at commit 9ff9a80, republished under its MIT licence (© LIDR-academy). 450 words, ~1,469 tokens.

Download SKILL.mdSave it as .claude/skills/run-parallel-tasks/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
run-parallel-tasks
description
Run N feature tasks in parallel, each in its own worktree, following the full specboot pipeline (enrich → new → ff → apply → verify). Stops after verify — no archive, no commit, no cleanup. Explicit task arguments override `parallel-tasks.md`; file is fallback only.
author
LIDR.co
version
1.1.0

run-parallel-tasks Skill

Resolves tasks from explicit command arguments first, otherwise reads parallel-tasks.md from the references subfolder. Then spins up one isolated agent per task and runs each through the full specboot pipeline without supervision.

Pipeline per agent: worktree → enrich-us → opsx:new → opsx:ff → opsx:apply → opsx:verify → stop

When to invoke

  • User says "run parallel-tasks.md"
  • User says "run the tasks" (when parallel-tasks.md exists in root)
  • User says "start the parallel tasks"
  • User provides explicit task arguments, for example: "run parallel tasks SCRUM-83 SCRUM-84"

Instructions

Step 1 — Resolve task source (arguments first, file fallback)

Parse user input for explicit task arguments after the trigger phrase (for example, in run parallel tasks SCRUM-83 SCRUM-84, the arguments are SCRUM-83 and SCRUM-84).

If one or more explicit task arguments are present:

  • Do not read parallel-tasks.md.
  • Build tasks directly from arguments.
  • For each argument <TASK_ID>, create:
    • name: <TASK_ID> lowercased to kebab-case
    • us: <TASK_ID> (treat as Jira source unless it clearly matches a file path)
  • Announce: "Using explicit task arguments (overriding file): name-1, name-2, ..."

If no explicit task arguments are present:

  • Read parallel-tasks.md from the references subfolder.
  • Extract every uncommented task block. A task block starts with ### Task and contains:
    • name: — kebab-case change name (required)
    • us: — source type: inline, a file path, or a Jira ticket ID (required)
    • description: — inline US text (required when us: inline)
  • Skip any task block wrapped in <!-- --> comments.
  • Announce: "Found N task(s): name-1, name-2, ..."
Show full SKILL.md (221 more words)Show less
Step 2 — Enrich US for each task

Always run enrich-us for every task, regardless of source format. Enrichment is mandatory.

First, resolve the raw US text by source:

  • us: inline — use the description: field as the raw US input.
  • us: <file-path> — read the file at that path as the raw US input.
  • us: <JIRA-ID> — pass the ticket ID directly to enrich-us; it fetches and enriches in one step.

Then, for inline and file-path sources, invoke the enrich-us skill passing the raw US text as input. Capture the full enriched output (with all technical detail, endpoints, files, test cases, NFRs) as the enriched US for this task.

Store each enriched US in memory before proceeding to Step 3.

Step 3 — Spawn one agent per task in parallel

Launch all agents simultaneously using the Agent tool with run_in_background: true.

Each agent must receive a fully self-contained prompt — agents start cold with no session memory. Use the template below, substituting:

  • {{CHANGE_NAME}} — the task name field
  • {{ENRICHED_US}} — the full resolved enriched US text from Step 2
  • {{PROJECT_ROOT}} — absolute path to the project root
  • {{BASE_BRANCH}} — current git branch (run git branch --show-current to get it)
Step 4 — Wait and report

When all background agents complete, print a summary table:

| Task | Worktree | Tasks | Verify | Blockers |
|------|----------|-------|--------|----------|
| name | .worktrees/name | N/N | PASSED / WARNINGS / CRITICAL | none / description |

If any agent reported CRITICAL issues or blockers, flag them clearly for the user.

Agent prompt template

You are implementing a feature for a full-stack FastAPI + React project.

Project root: {{PROJECT_ROOT}}
Base branch: {{BASE_BRANCH}}

Work exclusively inside the worktree you will create. Never modify the main checkout.

---

## Feature: {{CHANGE_NAME}}

## Enriched User Story

{{ENRICHED_US}}

---

## Pipeline — execute every step in order, do not skip any

### Step 1 — Create isolated worktree

Use the `using-git-worktrees` skill to create a worktree.
- Branch name: `feature/{{CHANGE_NAME}}-backend`
- Worktree path: `{{PROJECT_ROOT}}/.worktrees/{{CHANGE_NAME}}`
- Base: `{{BASE_BRANCH}}`

All subsequent steps run from inside this worktree.

### Step 2 — Create the OpenSpec change

Run from the worktree:
  openspec new change "{{CHANGE_NAME}}"

This scaffolds the change at `openspec/changes/{{CHANGE_NAME}}/`.

### Step 3 — Generate all artifacts (fast-forward)

Use the `opsx:ff` skill for change `{{CHANGE_NAME}}`.
Provide the Enriched User Story above as the primary source of truth for all artifacts.
Generate: proposal → design → specs → tasks in one pass.

### Step 4 — Implement all tasks

Use the `opsx:apply` skill for change `{{CHANGE_NAME}}`.
Work through every task in `tasks.md`. Do not stop until all tasks are marked `[x]`.

Rules:
- If a task requires a live DB or backend, start the required services first.
- After any CREATE/UPDATE/DELETE curl or Playwright test, restore the data state.
- Mark each task `[x]` immediately after completing it.

### Step 5 — Verify

Use the `opsx:verify` skill for change `{{CHANGE_NAME}}`.
Record all CRITICAL, WARNING, and SUGGESTION findings.

### Step 6 — Stop and report

Do NOT archive, commit, push, or remove the worktree.

Return this exact summary (fill in values):

TASK: {{CHANGE_NAME}}
WORKTREE: {{PROJECT_ROOT}}/.worktrees/{{CHANGE_NAME}}
TASKS_COMPLETE: N/N
VERIFY_RESULT: PASSED | WARNINGS | CRITICAL
ISSUES:
- <list any CRITICAL or WARNING issues, or "none">
BLOCKERS:
- <list anything that stopped progress, or "none">
TESTING REPORT SUMMARY:
- summary of the reports created
VERIFICATION SUMMARY:
- output from verify step as is

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

Files

SKILL.md and 1 other file (references) in ai-specs/skills/run-parallel-tasks of LIDR-academy/AI4Devs-LTI-extended.

  • SKILL.md
  • references/parallel-tasks.md

Open the folder on GitHubat commit 9ff9a80

Compare with similar skills

Run Parallel Tasks 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.

Run Parallel Tasks compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Run Parallel Tasks this skillLIDR-academy/AI4Devs-LTI-extended278—~1.5kAutomated safety check: PassMIT
Shep Workstreamsshep-ai/shep264—~2.5kAutomated safety check: PassMIT
Execute Fleetanombyte93/prd-taskmaster605—~2.2kAutomated safety check: NotesMIT
Vc Review Situationwithkynam/vibecode-pro-max-kit1.1k—~2.2kAutomated safety check: PassMIT
Project Session ManagerYeachan-Heo/oh-my-claudecode40k—~4kAutomated safety check: PassMIT
Acceptance Criteria Checkshopsys/shopsys350—~1kAutomated safety check: PassCustom licence

Similar skills

  • Shep Workstreams

    shep-ai/shep

    A skill your agent uses when a large body of work (a version milestone, an epic, a roadmap, a set of PRDs/design docs) needs to be broken into parallel workstreams and executed with the shep CLI.

    264 GitHub stars~2.5k tokensUpdated yesterday
    Product & Project ManagementAuto-check passed

  • Execute Fleet

    anombyte93/prd-taskmaster

    Phase execution skill for licensed Atlas Fleet runs. An agent skill from anombyte93/prd-taskmaster.

    605 GitHub stars~2.2k tokensUpdated 1 mo ago
    Product & Project ManagementAuto-check: notes

  • Vc Review Situation

    withkynam/vibecode-pro-max-kit

    A skill your agent uses when you need a read-only situation review and handoff summary of current branch state, local/remote refs, worktrees, active project plans, selected-plan hints, and suggested…

    1.1k GitHub stars~2.2k tokensUpdated 3 mo ago
    Product & Project ManagementAuto-check passed

  • Project Session Manager

    Yeachan-Heo/oh-my-claudecode

    Creates isolated git worktrees, with optional tmux sessions, for PR reviews, issue fixes and feature work across projects and repositories.

    40k GitHub stars~4k tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Acceptance Criteria Check

    shopsys/shopsys

    Checks a pull request against the acceptance criteria of its Jira issue by driving the deployed review environment with Playwright.

    350 GitHub stars~1k tokensUpdated today
    Product & Project ManagementAuto-check passed

  • Jira Natural Language Interface

    jjmartres/opencode

    Lets an agent view, create, update and transition Jira issues in natural language, automatically choosing between the jira CLI and Atlassian MCP tools.

    133 GitHub starsUsed in 3 repos~1.7k tokens
    Product & Project ManagementAuto-check passed

More from LIDR-academy/AI4Devs-LTI-extended

All 9 skills in this repo

  • Owasp Security Audit

    LIDR-academy/AI4Devs-LTI-extended

    A skill your agent uses when performing a cybersecurity audit, security review, OWASP Top 10 compliance check, vulnerability assessment, or preparing for a penetration test on a…

    278 GitHub stars~4.3k tokensUpdated 4 mo ago
    Auto-check: notes

  • Show Spec Working

    LIDR-academy/AI4Devs-LTI-extended

    A skill your agent uses when the user asks "show me X", "demo X", "walk me through X", "how X works" or requests a live feature demonstration from a spec, feature or ticket.

    278 GitHub stars~1.1k tokensUpdated 4 mo ago
    Auto-check passed

  • Sync Agent Symlinks

    LIDR-academy/AI4Devs-LTI-extended

    Analyze and synchronize agent skill exposure after ai-specs skill changes (additions, removals, renames).

    278 GitHub stars~1k tokensUpdated 4 mo ago
    Auto-check passed

  • Commit

    LIDR-academy/AI4Devs-LTI-extended

    Create focused commits and pull requests following repository standards.

    278 GitHub stars~1.5k tokensUpdated 4 mo ago
    Auto-check: notes

  • Explain

    LIDR-academy/AI4Devs-LTI-extended

    Teach underlying concepts with clear mental models to close skill gaps behind user questions.

    278 GitHub stars~1.3k tokensUpdated 4 mo ago
    Auto-check passed

  • Enrich Us

    LIDR-academy/AI4Devs-LTI-extended

    Analyze and enhance Jira user stories with complete, implementation-ready technical detail.

    278 GitHub stars~521 tokensUpdated 4 mo ago
    Auto-check passed

Works with

Categories

Questions about Run Parallel Tasks

What does Run Parallel Tasks do?

Run N feature tasks in parallel, each in its own worktree, following the full specboot pipeline (enrich → new → ff → apply → verify). Run Parallel Tasks is an agent skill from LIDR-academy/AI4Devs-LTI-extended. Run N feature tasks in parallel, each in its own worktree, following the full specboot pipeline (enrich → new → ff → apply → verify).

When should I use Run Parallel Tasks?

Run Parallel Tasks fits situations like: tasks that involve Git worktrees.

How do I install Run Parallel Tasks in Claude Code?

Run `npx skills add LIDR-academy/AI4Devs-LTI-extended --skill run-parallel-tasks -a claude-code`. Or copy the skill folder (ai-specs/skills/run-parallel-tasks in LIDR-academy/AI4Devs-LTI-extended) into .claude/skills/run-parallel-tasks in your project. Claude Code loads it when a task matches its description.

How do I install Run Parallel Tasks in Codex?

Run `npx skills add LIDR-academy/AI4Devs-LTI-extended --skill run-parallel-tasks -a codex`. Or copy the skill folder (ai-specs/skills/run-parallel-tasks in LIDR-academy/AI4Devs-LTI-extended) into .agents/skills/run-parallel-tasks in your project. Codex loads it when a task matches its description.

Can I use Run Parallel Tasks 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 LIDR-academy/AI4Devs-LTI-extended --skill run-parallel-tasks -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/run-parallel-tasks, .gemini/skills/run-parallel-tasks, .github/skills/run-parallel-tasks and .opencode/skills/run-parallel-tasks in your project.

What does Run Parallel Tasks need to run?

Going by SKILL.md and its folder, Run Parallel Tasks needs the command-line tools its instructions call (git).

Does Run Parallel Tasks access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Run Parallel Tasks 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 Run Parallel Tasks use?

Run Parallel Tasks 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 Run Parallel Tasks use?

About 1.5k tokens (SKILL.md is roughly 5.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 357 tokens, read only when the agent opens those files.

What are the alternatives to Run Parallel Tasks?

Skills that share tags, products or a category with Run Parallel Tasks: Shep Workstreams (shep-ai/shep, 264 stars), Execute Fleet (anombyte93/prd-taskmaster, 605 stars), Vc Review Situation (withkynam/vibecode-pro-max-kit, 1.1k stars) and Project Session Manager (Yeachan-Heo/oh-my-claudecode, 40k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Run Parallel Tasks?

LIDR-academy (a GitHub organization) maintains it in LIDR-academy/AI4Devs-LTI-extended, which has 278 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on June 9, 2026.

Source: LIDR-academy/AI4Devs-LTI-extended on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.