[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-79894":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":15,"stars7d":16,"stars30d":17,"stars90d":15,"forks30d":15,"starsTrendScore":15,"compositeScore":18,"rankGlobal":9,"rankLanguage":9,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":9,"pushedAt":9,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":15,"starSnapshotCount":15,"syncStatus":13,"lastSyncTime":27,"discoverSource":28},79894,"claude-goal","jthack\u002Fclaude-goal","jthack","Codex-style \u002Fgoal command for Claude Code",null,"Python",101,13,2,1,0,3,6,45.54,"MIT License",false,"main",true,[],"2026-06-12 04:01:25","# claude-goal\n\nA Codex-style `\u002Fgoal` command for Claude Code.\n\nIt gives Claude Code a persistent local goal state, Codex-inspired continuation instructions, pause\u002Fresume\u002Fclear\u002Fstatus controls, completion-audit guardrails, and a Stop hook that keeps Claude working while a goal is active.\n\n## Install\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fjthack\u002Fclaude-goal.git\ncd claude-goal\n.\u002Finstall.sh\n```\n\nThis installs:\n\n- `~\u002F.claude\u002Fskills\u002Fgoal` as a symlink to this repo's `goal\u002F` directory\n- a user-level Claude Code `Stop` hook in `~\u002F.claude\u002Fsettings.json`\n\nThe `goal\u002F` directory is the Claude skill package. It contains `SKILL.md`, `scripts\u002Fclaude_goal.py`, and reference notes.\n\nState is stored at:\n\n```text\n~\u002F.claude\u002Fgoal\u002Fgoals.sqlite\n```\n\n## Usage\n\n```text\n\u002Fgoal find and fix the flaky auth tests\n\u002Fgoal --tokens 250K do deep research and build the full prototype\n\u002Fgoal\n\u002Fgoal status\n\u002Fgoal pause\n\u002Fgoal resume\n\u002Fgoal clear\n```\n\nWhen a goal is active, the command returns a continuation prompt that wraps the goal text in `\u003Cobjective>` and requires a completion audit before marking the goal complete.\n\n## Notes\n\nClaude Code custom skills do not currently expose reliable live per-turn token usage to markdown commands. Token budgets are therefore stored and displayed as soft budgets. Elapsed-time tracking is local and persistent.\n\nThe Stop hook blocks Claude from stopping while the current goal is active. It stops blocking when you run `\u002Fgoal pause`, `\u002Fgoal clear`, or `\u002Fgoal complete`.\n\nBy default, the runaway guard allows up to 500 Stop-hook continuations for a single active goal. That high default is intentional: `\u002Fgoal` is meant for long-running work where Claude may need many turns to finish. If you want a stricter cap, set `CLAUDE_GOAL_MAX_STOP_CONTINUES` before launching Claude Code:\n\n```bash\nexport CLAUDE_GOAL_MAX_STOP_CONTINUES=50\n```\n\n## Test\n\n```bash\npython3 -m pytest tests\n```\n","claude-goal 项目为 Claude Code 提供了一个类似于 Codex 的 `\u002Fgoal` 命令。其核心功能包括持久化的本地目标状态管理、受 Codex 启发的连续指令、暂停\u002F恢复\u002F清除\u002F状态控制、完成审计防护以及在目标活动期间阻止 Claude 停止工作的 Stop 钩子。该项目使用 Python 开发，适合需要长时间运行且涉及多步骤任务处理的场景，如深入研究和原型构建等复杂开发工作。通过简单的命令行操作即可实现对目标的有效管理和追踪。","2026-06-11 03:58:26","CREATED_QUERY"]