{
  "markdown": "<p align=\"center\">\n  <a href=\"https://claude-automata.clomia.com/\"><img src=\"https://raw.githubusercontent.com/clomia/claude-automata/main/site/assets/banner.png\" alt=\"claude-automata: 24/7 full self-driving for Claude Code\" width=\"840\"></a>\n</p>\n\n<p align=\"center\"><strong>Your agent stops when it thinks it's done. This one stops when it's actually done.</strong></p>\n\n<p align=\"center\">\n  An agent environment for Claude Code, modeled on human memory.<br>\n  Hand it months of work and rest: it finishes in days.\n</p>\n\n<p align=\"center\"><a href=\"https://claude-automata.clomia.com/\"><strong>See more</strong></a></p>\n\n<p align=\"center\"><sub>This README is the summary. The site walks the whole system, animated.</sub></p>\n\n<p align=\"center\">\n  <a href=\"https://pypi.org/project/claude-automata/\"><img src=\"https://img.shields.io/pypi/v/claude-automata?style=flat&color=f54e00\" alt=\"PyPI\"></a>\n  <a href=\"https://github.com/clomia/claude-automata/blob/main/LICENSE\"><img src=\"https://img.shields.io/pypi/l/claude-automata?style=flat&color=447e48\" alt=\"License\"></a>\n</p>\n\nEnglish | [한국어](https://github.com/clomia/claude-automata/blob/main/README.ko.md)\n\n---\n\nClaude Code ends its turn the moment it believes it's finished, and loses the detail at the next compaction. claude-automata rebuilds it around the way human memory works:\n\n| Plugin | Memory role |\n|---|---|\n| **ploop** | working memory: a single session sails autonomously for days; an independent advisor judges every completion claim |\n| **tx** | consolidation: the screening gate into long-term memory (independent verify, CI, squash merge) |\n| **refine** | re-grounding: large-scale workflows that eliminate technical debt |\n\nLong-term memory is the repository's own git-tracked text, not a database. Recall is grep. Whatever never passes the gate dies with the loop, on purpose.\n\n## Getting started\n\nEnter the following instruction in Claude Code:\n\n```\nRead https://raw.githubusercontent.com/clomia/claude-automata/refs/heads/main/INSTALL.md with curl -sSL and install claude-automata.\n```\n\nYour agent reads [INSTALL.md](https://github.com/clomia/claude-automata/blob/main/INSTALL.md) and installs claude-automata. Needs [Claude Code](https://claude.com/claude-code) and [uv](https://docs.astral.sh/uv/getting-started/installation/) on POSIX (macOS / Linux / WSL).\n\n## Operating the loop\n\n```\n/ploop:define-mission          # an agent interviews you, interprets your intent, and writes the anchor\n/ploop:launch [anchor text]    # hand it to the loop in a fresh session\n```\n\nDeclare the mission done and a hook blocks the stop, demanding an audit by an independent advisor with access to the whole story; the loop ends only when the advisor certifies completion.\n\n```\nagent   › Mission accomplished. Stopping.\nadvisor › Not yet. The mobile layout was never measured. …resuming.\n```\n\nSix rounds later the advisor ends the turn. [Watch a full round on the site.](https://claude-automata.clomia.com/#advisor) The anchor survives every auto-compaction, and the loop fully complies with your subscription plan's terms of service.\n\n<details>\n<summary><strong>Pause, resume, observe</strong></summary>\n\n<br>\n\n- init sets Auto-Compact; keep it on. For unattended runs, set `askUserQuestionTimeout` so an unanswered question can't park the loop.\n- `/ploop:off` pauses the loop. `/ploop:on` resumes or restores it, even after an accidental ESC, an API error, or a session limit (interrupt with ESC first if a turn is running). Nothing else stops the loop.\n- `/ploop:docent` reports the loop's progress. Run it in a **separate session, same directory**: questions go to the docent, interventions go straight to the loop session.\n\n</details>\n\n## Change as transactions\n\nAgents drive tx on their own. Every change lands as one verified, CI-green squash merge behind an integrity boundary, and tx blocks writes to the base branch in between. You review merged results, not work in progress.\n\n## Keep the repository lean\n\n```\n/refine:code [focus] · /refine:docs [focus]\n```\n\nRepository-wide workflows that clear technical debt: code architecture and doc-to-code alignment. They touch only the representation layer and never change behavior. A run can sweep the whole codebase and take ten hours or more; watch it with `/workflows`.\n\n---\n\n<p align=\"center\"><a href=\"https://claude-automata.clomia.com/\"><strong>See more</strong></a></p>\n\nApache-2.0 · Recursive self-improvement: claude-automata is developed inside claude-automata. A Claude Code agent running this environment authored every contribution in this repository.\n",
  "bytes": 4575,
  "sha": "0b85badc07e4ed6fd5526e1f6bcaf26b01781fa6870749217b9d18d56467ae4b",
  "repo_slug": "clomia/claude-automata",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_clomia_claude_automata_parallax_8e6417a1/readme"
}