[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1873":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":14,"subscribersCount":14,"size":14,"stars1d":15,"stars7d":15,"stars30d":15,"stars90d":14,"forks30d":14,"starsTrendScore":12,"compositeScore":16,"rankGlobal":9,"rankLanguage":9,"license":17,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":20,"hasPages":18,"topics":21,"createdAt":9,"pushedAt":9,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":14,"starSnapshotCount":14,"syncStatus":15,"lastSyncTime":33,"discoverSource":34},1873,"no-no-debug","summerliuuu\u002Fno-no-debug","summerliuuu","No-No Debug — Self-evolution system for AI coding assistants. 10 minutes writing code, 2 hours debugging? This skill makes your AI remember all its bugs.",null,"Python",152,6,3,0,2,47.74,"MIT License",false,"main",true,[22,23,24,25,26,27,28,29],"ai-agent","claude","claude-code","debug","error-tracking","self-evolution","self-improvement","skill","2026-06-12 04:00:11","# no-no-debug\n\nA self-evolution system for AI coding assistants\n\n> **v1.3.0 — periodic review now actually works.** Earlier versions promised\n> \"auto-triggers every 3 days\" but had no enforcement mechanism — the review\n> only ran when manually invoked, and the tracker went stale silently. v1.3.0\n> ships `hooks\u002Freview_reminder.py`, a `UserPromptSubmit` hook that checks\n> the tracker on each message and reminds the AI to run the review when it's\n> overdue. See [CHANGELOG.md](.\u002FCHANGELOG.md) for the one-step migration\n> from v1.2.0.\n\n## What problem does this solve?\n\n10 minutes writing code, 2 hours debugging.\n\nThis skill fills the gap in AI's cross-session error memory, delivering three things:\n\n1. Dramatically less debug time\n2. Higher code quality\n3. A self-evolution feedback loop\n\nThe longer you use it, the fewer mistakes get repeated.\n\n## How it works\n\n### 1. Real-time Logging (automatic)\nWhen the AI is corrected, code errors occur, deploys fail, or tests don't pass — it automatically appends to a local error_log.md with a timestamp. No need to say \"write that down.\"\n\n### 2. Three Gates (on every code change)\nBefore the change: what does this affect?\nAfter the change: did you actually verify it?\nBefore deploying: did you test with a non-admin account?\n\nRuns silently. No output when all gates pass.\n\n### 3. Periodic Review (auto-triggers every 3 days)\nReads error_log.md, categorizes by dimension, updates error_tracker.md, outputs an evolution report.\nReview frequency is configurable: 1 day \u002F 3 days (default) \u002F 7 days.\n\n### 4. Rule Accumulation\nNew error type → automatically creates a prevention rule.\nRepeated offense → counter increments, rule strengthens.\n4 consecutive clean periods → marked as cured.\n\nRules persist across sessions — nothing gets lost.\n\n### 5. Confirmation Gate\nThe following situations require user confirmation before proceeding:\n- New feature development (not a bug fix)\n- Changes involving databases, environments, or deployments\n- Publishing to external platforms\n- When the user raises a new idea or new direction mid-task\n\n### 6. Auto Hooks\nAutomatically configures Claude Code hooks on install:\n- Command errors → auto-logged to error_log.md\n- After editing a file → auto-reminder to verify\n- When user corrects the AI → correction content auto-logged\n- Review overdue → auto-reminder to start Mechanism 3 review\n\n## Tracked Dimensions\n\n| Dimension | What it tracks |\n|-----------|---------------|\n| Data Accuracy | Do displayed numbers\u002Fformulas match actual code |\n| Environment Safety | Did config changes break login or the database |\n| Foresight | Were permission, migration, or cache issues caught before deploy |\n| User Perspective | Does the feature work end-to-end from the user's account |\n| Verification | Was there a real end-to-end test after the fix |\n| Memory Consistency | Did the AI read existing records instead of asking again |\n| Tool Judgment | Did failing tools get swapped out promptly |\n| Review Completeness | Were reviews and summaries thorough with nothing missed |\n| Operational Precision | Did changes produce unintended side effects |\n| Check Before Doing | With unfamiliar tools\u002Fversions, was documentation checked first |\n| Conciseness | Was a 3-line solution turned into 300 lines |\n| Regression Awareness | Did fixing one bug introduce a new one |\n| Style Consistency | Does new code follow the project's existing style and architecture |\n| Independent Judgment | When the user's premise is wrong, does the AI push back instead of blindly executing |\n| Real-env Verification | Was the fix validated with the real production command, not a sandbox\u002Ftest harness |\n| Cross-agent Trust | Did the AI re-verify another agent's \"pass\" report instead of trusting it blindly |\n| Dumb things humans will do | Not yet committed, but inevitable |\n| Dumb things AI will do | Same, but the AI edition |\n\n## Real-world data\n\n| Period | Errors | Notes |\n|--------|--------|-------|\n| Week 1 | 29 | Baseline, 10 dimensions identified |\n| Week 2 | 6 | Rules starting to take effect |\n| Week 3 | ~0 | Still running |\n\n## Installation\n\n### Claude Code (recommended — full experience with auto hooks)\n```bash\nclaude skill add summerliuuu\u002Fno-no-debug\n```\nZero configuration. Tracking files and hooks are initialized automatically on first run.\n\nIf you're installing manually or upgrading from v1.1.0, also copy the hook scripts:\n\n```bash\nmkdir -p ~\u002F.claude\u002Fhooks\ncp hooks\u002Fuser_prompt_filter.py ~\u002F.claude\u002Fhooks\u002F\ncp hooks\u002Fpost_tool_failure.sh ~\u002F.claude\u002Fhooks\u002F\ncp hooks\u002Freview_reminder.py ~\u002F.claude\u002Fhooks\u002F\nchmod +x ~\u002F.claude\u002Fhooks\u002Fpost_tool_failure.sh\n```\n\nThen merge the `PostToolUseFailure` \u002F `UserPromptSubmit` blocks from `SKILL.md` → Mechanism 6 into `~\u002F.claude\u002Fsettings.json`.\n\n**Prerequisites.** The shipped hooks use `python3` (UserPromptSubmit filter) and `jq` (PostToolUseFailure parser). Both ship by default on recent macOS and most Linux distros; if either is missing the corresponding hook degrades gracefully — `python3` missing means the correction detector silently no-ops, `jq` missing writes a one-time `jq_not_installed` marker to the log. Install with `brew install jq` (macOS) or your package manager if you want full coverage.\n\n### ChatGPT \u002F Cursor \u002F Copilot \u002F Other AI assistants\n1. Copy the contents of [SKILL.md](.\u002FSKILL.md)\n2. Paste it into your system prompt or custom instructions\n3. Core features (3-gate checkpoint, periodic review, rule accumulation) work immediately\n4. Auto hooks are Claude Code only — other tools rely on the AI following the rules in SKILL.md\n\n## Compatibility\n\n- **Claude Code** — full experience with auto hooks and one-line install\n- **Other AI coding assistants** (ChatGPT, Cursor, Copilot, etc.) — paste SKILL.md content into your system prompt to use core features\n- Any project, any language\n- Pairs well with claude-mem (for session search)\n\n## ⭐\n\nThank you for starring this project during installation.\n\n📕 [Xiaohongshu @Summer的AI日常](https:\u002F\u002Fxhslink.com\u002Fm\u002F6fWf0EN8e6) — AI coding in practice\n\n---\n\n🇨🇳 [中文版](.\u002FREADME.zh-CN.md)\n\n## License\n\nMIT — [@summerliuuu](https:\u002F\u002Fgithub.com\u002Fsummerliuuu)\n","No-No Debug 是一个专为AI编码助手设计的自我进化系统，旨在减少调试时间、提高代码质量。其核心功能包括实时错误记录、三重验证门、定期回顾（默认每3天自动触发）、规则累积和用户确认门等。通过这些机制，该系统能够帮助AI记住跨会话的错误，并逐渐减少重复错误的发生。项目采用Python语言编写，适合那些希望提升开发效率、降低调试成本且追求高质量代码输出的开发者或团队使用。","2026-06-11 02:46:35","CREATED_QUERY"]