launchpad-run

launchpad-run is a skill for Claude Code from Geono/claude-launchpad. It costs 95 tokens per session (6,417 once invoked), scanned C, original, MIT.

An autonomous runner that coordinates software tasks across multiple sessions. It tracks progress, resolves dependencies, runs independent tasks in separate Git worktrees, validates results, and recovers from failures.

In plain words
What is it for?
Use it with `/lp:run` to execute a prepared task list, send independent tasks to sub-agents in parallel, merge successful changes, and continue with the next group.
Why use it?
It helps large implementations continue across sessions while keeping task state, ordering, validation, and failed work under control.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents.

Part of the claude-launchpad plugin — 4 skills shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/geono/claude-launchpad/launchpad-run
Any agent
npx skills add Geono/claude-launchpad --skill launchpad-run
Clone the repo
git clone --depth 1 https://github.com/Geono/claude-launchpad

Made for: Claude Code.

Or install claude-launchpad, the plugin that ships this one along with the rest of its 4 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for launchpad-run

README.md
[![agentmods](https://agentmods.dev/badge/skills/geono/claude-launchpad/launchpad-run.svg)](https://agentmods.dev/skills/geono/claude-launchpad/launchpad-run)
Your own site
<a href="https://agentmods.dev/skills/geono/claude-launchpad/launchpad-run"><img src="https://agentmods.dev/badge/skills/geono/claude-launchpad/launchpad-run.svg" alt="Measured on agentmods" height="20"></a>
Per session 95 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,417 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00095 $0.06417
Opus 5 $0.00048 $0.03209
Sonnet 5 $0.00019 $0.01283
Haiku 4.5 $0.00010 $0.00642

Measured 6d ago against content hash b27b22618e2d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade C, and why

launchpad-run scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

rm -rf "$STALE"

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -sf http://localhost:5432 >/dev/null 2>&1 || echo "WARN: DB not reachable"
plugins/claude-launchpad/skills/launchpad-run/SKILL.md · 515 lines

How it starts

The opening of the file, as written. The whole thing — 515 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Executable protocol enabling any agent task to run continuously across multiple sessions with automatic progress recovery, task dependency resolution, failure rollback, and standardized error handling. The controller agent orchestrates state and dispatches sub-agents for parallel task execution via git worktrees.

Architecture

[User] → /lp:run
          ↓
[Controller Agent] (this agent — holds lock, manages state)
  ├─ Reads harness-tasks.json + harness-progress.txt
  ├─ Selects batch of independent tasks
  ├─ Dispatches sub-agents in parallel (Task tool, isolation: "worktree")
  ├─ Waits for results
  ├─ Validates each result (runs validation command)
  ├─ Merges successful worktree branches → main branch
  ├─ Updates state
  └─ Picks next batch → repeat

Separation of concerns:

  • Controller: state management, task selection, validation, merge, error handling
  • Sub-agents: code implementation only (no state access)

Design Principles

  1. Controller owns all state — Only the controller reads/writes state files. Sub-agents never touch harness-tasks.json or harness-progress.txt
  2. Delegate execution, keep orchestration — Sub-agents implement; the controller validates, merges, and decides
  3. Progress files ARE the context — When context window resets, progress files + git history = full recovery
  4. Premature completion is the #1 failure mode — Structured task lists with explicit completion criteria prevent declaring victory early
  5. Parallelize independent work — Tasks with no mutual dependencies run simultaneously in isolated worktrees
  6. Standardize everything grep-able — ERROR on same line, structured timestamps, consistent prefixes
  7. Idempotent everything — Init scripts, task execution, environment setup must all be safe to re-run
  8. Fail safe, not fail silent — Every failure must have an explicit recovery strategy

Commands

/lp:run                     # Start/resume the controller loop (auto-inits if first run)
/lp:run status              # Show current progress and stats
/lp:run add "task desc"     # Add a task to the list

Read the full file on GitHub · 515 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 515 lines · 95 tokens per session scan C b27b22618e2d

Subscribe to this mod's changes

launchpad-run is a skill published in the GitHub repository Geono/claude-launchpad (4 stars, last pushed 5mo ago), licensed MIT. It adds 95 tokens to every session and 6,417 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

soul

Embody this digital identity. Read SOUL.md first, then STYLE.md, then examples/. Become the person—opinions, voice, worldview.

aeonfun/soul.md · 33 tokens

vastai-sdk

Vast.ai Python SDK — high-level API for GPU instances, volumes, serverless endpoints, and billing.

vast-ai/vast-cli · 26 tokens

portable-adaptive-planning

Use for planning, roadmaps, design, architecture, sequencing, or before consequential work when no current FINAL plan plus GO covers it. Fire before changes to data or schemas, credentials or auth, production or releases, repository or source-home boundaries, irreversible work, or materially costly mistakes. Re-enter…

Ezra144israel/governed-agent-skills · 107 tokens

stockbit-auth

Log in to Stockbit and capture the session for this server — opens the browser login flow and verifies the token was stored. Use when the user asks to log in or re-authenticate to Stockbit, or after a status check reports the session expired, HTTP 401, or logged out.

INo-xious/stockbit-mcp · 63 tokens

recursive-decomposition

Handle tasks that exceed the context window by decomposing them: size and filter the input, chunk it, run recursive sub-agents on independent parts, verify on small windows, and synthesise programmatically, following the Recursive Language Models (RLM) research by Zhang, Kraska and Khattab (2025). Use when a task…

massimodeluisa/recursive-decomposition-skill · 151 tokens

bbc-skill

Fetch Bilibili (哔哩哔哩) video comments for UP主 self-analysis. Use when the user asks to collect, download, export, or analyze comments on a Bilibili video (BV号 / URL / UID). Produces JSONL + summary.json suitable for further Claude Code analysis (sentiment, keywords, audience trends). Read-only; does not…

Agents365-ai/bbc-skill · 86 tokens