#291291st of 455 in Skills, plugins and prompts
autoprompt-skill
Plan, build, check and repair workflow layered on eleven coding agents
- Stars
- 1.2k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- JavaScript
317 on 19 Aug 20261,228 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Autoprompt is an installable skill and CLI that wraps coding agents such as Claude Code, Codex, OpenCode and Grok Build in a plan, build, independent-check and repair loop. It launches with `autoprompt activate PROVIDER -- "<goal>"`, with work paths (auto, direct, light, roadmap) and concurrency limits. On Terminal-Bench 2.1 with OpenCode it solved 73/89 tasks versus 60/89 without it.
Who it is for: Developers who want a verify-and-repair loop around their existing coding agent
Strengths
- Same controls across 11 providers, including Claude Code, Codex, OpenCode and VS Code
- Separate checker agents verify work instead of the agent approving itself
- Concurrency modes: tokensaver caps at six subagents, wide, or custom --max-subs
- Reports 73/89 vs 60/89 on Terminal-Bench 2.1 with OpenCode, with setup notes
Weaknesses
- Costs about 3x time and 2x tokens; figures are estimates, logs not retained
- Benchmark covers one agent (OpenCode) and is labeled version 1
- Needs Node 20+, Python 3.11+ with PyYAML, and Bash 4.3+ on macOS or Linux
- Support table tested on Linux only; native Windows Codex needs admin controller
What it needs
- no GPU
- Needs Node.js 20+, Python 3.11+, PyYAML, Bash 4.3+ (macOS/Linux), Git (GitHub checkout only)
- Models: Claude Code, Codex, OpenCode, Kilo Code, VS Code
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Skills, plugins and prompts
See all 455| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |