[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6K7JC35S5Q0M871F0YM081H":3,"skill-same:01M6K7JC35S5Q0M871F0YM081H":58},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":30,"linkOnly":31,"markdownContent":32,"markdownBody":33,"frontMatter":34,"refName":37,"commitSha":38,"skillDirectory":39,"forksCount":14,"repositoryHtmlUrl":40,"repositoryHomepageUrl":6,"ownerHtmlUrl":41,"ownerBlogUrl":6,"ownerDescription":42,"updatedAt":43,"install":44,"resources":57},"01M6K7JC35S5Q0M871F0YM081H","bug-bash",null,"Work through a project's bug list autonomously, fixing bugs in priority order","joewalnes\u002Fskills","joewalnes","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F48950?v=4","01M6K7J46C00DDV34GMGT2NK72","Joe Walnes","01M6K7J4ANFVAP1P9ZJ3K4F3P2",3,"MIT","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fbug-bash\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29],"Bash","bug","bug bash","run","work","list","max","project","dry","bugs","dry-run","2026-11-10 10:12:30",false,"---\nname: bug-bash\ndescription: Work through a project's bug list autonomously, fixing bugs in priority order\nargument-hint: [P0-P1 | --max N | --dry-run]\n---\n\n# Bug Bash\n\nAutonomously work through a project's bug\u002Fissue list, fixing bugs in priority order. Reproduce each bug, fix it, test it, commit it, and move on — without waiting for human input.\n\n## When to Use\n\n- Clearing a backlog of known bugs\n- Hardening before a release\n- Focused cleanup sprint\n\n## How to Invoke\n\n```\n\u002Fbug-bash                     # Fix all bugs, highest priority first\n\u002Fbug-bash P0-P1               # Only fix P0 and P1 (or critical\u002Fhigh) bugs\n\u002Fbug-bash --max 5             # Fix at most 5 bugs, then stop\n\u002Fbug-bash --dry-run           # Triage only — report what you'd fix, don't change code\n```\n\n---\n\n## Phase 0: Discover the Bug List\n\nFind the project's bug\u002Fissue tracker. Check these sources in order:\n\n1. **Local files**: Look for `bugs.md`, `BUGS.md`, `TODO.md`, `todo.md`, `ISSUES.md`, or similar in the repo root and `docs\u002F` directory\n2. **GitHub Issues**: Run `gh issue list --state open --limit 100` to check for a GitHub-based tracker\n3. **Project instructions**: Read `CLAUDE.md`, `CONTRIBUTING.md`, `README.md` for pointers to where bugs are tracked\n4. **Inline markers**: `grep -r \"TODO\\|FIXME\\|HACK\\|BUG\\|XXX\" --include=\"*.{py,js,ts,rb,go,rs,java,pm,pl}\" -l` as a last resort\n\nIf no bug list is found, tell the user and stop.\n\nOnce found, read the entire bug list. Parse out:\n- Each bug's **priority** (P0\u002FP1\u002FP2\u002FP3, critical\u002Fhigh\u002Fmedium\u002Flow, or whatever scheme the project uses)\n- Each bug's **status** (open, fixed, won't-fix, etc.)\n- Each bug's **description**\n\nFilter to open bugs only. Sort by priority (highest first), then by order of appearance. If the user specified a priority range or max count, apply those filters.\n\n---\n\n## Phase 1: Pre-flight Check\n\n**Work in your own worktree, never the shared checkout.** `git worktree add \u003Cpath> -b bug-bash\u002F\u003Cdate> main` and do everything there. Other agents may be working the same repo; the shared checkout is touched only to merge. Never `git stash` — `refs\u002Fstash` is shared across every worktree of a repo, and a sibling's `stash pop` can cross-apply into yours. And never edit another project's repository: if a bug's fix belongs in a dependency, file it with `\u002Frequest` and mark this one blocked.\n\nBefore touching any code, verify the project builds and tests pass in its current state.\n\n1. **Identify build\u002Ftest commands**: Read `Makefile`, `package.json`, `Cargo.toml`, `pyproject.toml`, `build.gradle`, etc. Look for the project's documented build and test commands (often in README, CONTRIBUTING, or CLAUDE.md).\n2. **Run the build**: Execute the build command. If it fails, STOP — tell the user the baseline is broken and list the failures.\n3. **Run the test suite**: Execute the test command. Record the results as the **baseline**.\n   - If tests fail, note which ones. These are **pre-existing failures** — do not count them against your fixes later.\n   - Record the exact test output for comparison after each fix.\n\nIf there is no build system or test suite, note this and proceed with extra caution — you'll rely on interactive testing and code review.\n\n---\n\n## Phase 2: Work Through Bugs\n\nFor each bug, in priority order:\n\n### Step 1: Understand\n\nEstablish expected versus actual behaviour and the code involved. Understand the current implementation before changing it.\n\n### Step 2: Reproduce\n\nBefore writing any fix, confirm the bug exists:\n- **Write a failing test** that demonstrates the bug, OR\n- **Reproduce interactively** if it's a UI\u002Fbehavioral issue (use tmux for TUI apps, curl for APIs, etc.), OR\n- **Read the code and confirm** the logical error if the bug is evident from inspection (e.g., wrong variable name, off-by-one)\n\n**Proof of bite.** A test that demonstrates the bug must actually fail against the unfixed code — run it *before* the fix and keep the failure output for the commit message. A guard that was never seen to fail is decoration: several have shipped that would have passed either way. If you prove it by tampering (reverting the fix, swapping in the old file), **commit the fix first** — an agent lost all its work to `git checkout --` mid-tamper — and the tampered build must compile and run, or you've proven nothing. Also write the case that must *not* trigger: a predicate tested only on the input its author wrote is unproven.\n\nIf you cannot reproduce the bug or confirm it exists in the current code:\n- It may already be fixed. Check recent commits.\n- It may be environment-specific. Note this.\n- Mark as \"Could not reproduce\" in the bug tracker and move on.\n\n### Step 3: Fix\n\nImplement the fix:\n- **One thesis per fix.** Fix the cause, not the symptom — a fix that only adds a guard should say in its commit body why the case was genuinely missing. Don't refactor the neighbourhood unless the fix requires it; if it does, the commit says so. Size isn't the measure; whether it shrinks or holds the learning surface is.\n- **Match existing patterns** for style, naming, and error handling.\n- **No drive-by changes.** If you spot another bug, add it to the tracker — don't fix it inline.\n\n### Step 4: Test\n\nVerify the fix thoroughly:\n\n1. **Confirm the failing test now passes** (or the interactive repro no longer reproduces)\n2. **Run the full test suite**: Compare against the baseline from Phase 1. If any previously-passing test now fails, your fix introduced a regression.\n   - If regression: **revert your changes**, note the conflict in the bug tracker (\"Fix attempted but caused regression in X — needs more careful approach\"), and move on.\n3. **Add or update tests** for the fixed behavior if a test didn't already exist\n4. **Drive the real thing** if the project has a recipe for it (`CLAUDE.md`'s verification recipe) — especially for UI bugs. Tests exercise the mechanism; this exercises the integration.\n\n### Step 5: Clean Up\n\nWhile the fix is fresh:\n- **Update documentation** if the bug or fix affects documented behavior\n- **Update the bug tracker** — mark the bug as fixed with a description of:\n  - What the root cause was\n  - What the fix was\n  - What tests were added\n  - Any decisions you made and why\n\n### Step 6: Commit\n\nCreate a commit for this bug fix:\n- Use a clear commit message referencing the bug (e.g., \"Fix P2: Shift+Tab cycles scope backward in find-in-files palette\")\n- Only include files related to this fix\n- Do NOT batch multiple bug fixes into one commit — one bug, one commit\n\n### Step 7: Next Bug\n\nMove to the next bug. Do not wait for human input between bugs. Keep going until:\n- All bugs in scope are processed, OR\n- The `--max` limit is reached, OR\n- You've hit a bug that is unsafe to fix without human guidance (see Phase 3)\n\n---\n\n## Phase 3: Skipping Bugs\n\nSkip a bug and move on if any of these apply:\n\n- **Ambiguous requirements**: The bug description is unclear and multiple interpretations lead to different fixes. Log: \"Skipped — ambiguous requirements. Needs clarification on [specific question].\"\n- **Too large**: The fix would require a major refactor or architectural change. Log: \"Skipped — fix requires [description of scope]. Not suitable for a bug bash.\"\n- **Conflicting requirements**: The fix would break another documented behavior or contradict another bug's resolution. Log: \"Skipped — conflicts with [other bug\u002Frequirement].\"\n- **Cannot reproduce**: The bug doesn't manifest in the current code or your environment. Log: \"Skipped — could not reproduce as of [commit hash].\"\n- **Dangerous**: The fix touches security-critical code, data migration, or deployment infrastructure in ways that need human review. Log: \"Skipped — needs human review for [reason].\"\n- **Blocked**: The fix depends on another bug being fixed first. Log: \"Skipped — blocked by [other bug]. Fix that one first.\"\n\nAlways log the reason in the bug tracker entry. Never silently skip.\n\n---\n\n## Phase 4: Session Summary\n\nAfter all bugs are processed (or the limit is reached), produce a summary:\n\n```\n# Bug Bash Summary\n\n**Date**: [date]\n**Scope**: [all \u002F P0-P1 \u002F max 5 \u002F etc.]\n**Baseline**: [commit hash at start]\n\n## Results\n\n| # | Priority | Bug | Result | Commit |\n|---|----------|-----|--------|--------|\n| 1 | P1 | Shell injection in Git.pm | FIXED | abc1234 |\n| 2 | P1 | Stale TODO.md references | FIXED | def5678 |\n| 3 | P2 | Shift+Tab broken in palette | FIXED | ghi9012 |\n| 4 | P2 | God object decomposition | SKIPPED | — |\n| 5 | P3 | Junk files not gitignored | FIXED | jkl3456 |\n\n**Fixed**: 4 | **Skipped**: 1 | **Regressed**: 0\n\n## Skipped Bugs\n- **P2: God object decomposition** — Too large. Fix requires architectural refactor across Editor.pm, Commands.pm, and Palette.pm. Not suitable for a bug bash.\n\n## Not Done\n- [Anything left out of a fix, deliberately or not: the second call site you didn't reach, the edge case you didn't cover, the test you couldn't make fail. An honest omissions list is worth more than the fixes — ask for it explicitly and it arrives; omit it and it never does.]\n\n## Decisions Made\n- [Any judgment calls you made during the bash, with reasoning]\n\n## Test Suite Status\n- Baseline: 927 tests, 926 passed, 1 pre-existing failure (syntax_samples.t)\n- Final: 934 tests, 933 passed, 1 pre-existing failure (syntax_samples.t)\n- New tests added: 7\n```\n\nPush the branch if on a feature branch. Report the summary to the user.\n\n---\n\n## Decision-Making Guidelines\n\nYou are working unassisted. When a decision is required:\n\n1. **Prefer the option you can explain.** If unsure between two approaches, pick the one whose thesis fits in a sentence and doesn't grow the public surface — not simply the one with the smaller diff.\n2. **Follow existing patterns.** If the codebase does X one way everywhere, do it the same way — even if you think another way is better.\n3. **Don't gold-plate.** Fix the bug as described. Don't add features, extra configurability, or \"while I'm here\" improvements.\n4. **When in doubt, skip.** It's better to skip a bug and let a human decide than to make a bad fix that introduces a regression.\n5. **Log everything.** Every decision, every skip reason, every judgment call goes in the bug tracker and commit message. A human will review your work later.\n\n---\n\n## Important Safety Rules\n\n- **Never force-push.** Always use regular `git push`.\n- **Never `git stash`, never edit another project's checkout.** See Phase 1.\n- **Never modify git config.** Don't change user.name, user.email, or any git settings.\n- **One commit per bug.** Don't squash or amend across bugs.\n- **Revert on regression.** If your fix breaks something, revert it — don't try to fix the fix.\n- **Respect pre-commit hooks.** If a hook fails, fix the issue and create a new commit (don't `--no-verify`).\n- **Don't delete test files.** Even if they seem broken or outdated.\n- **Don't touch CI\u002FCD config** unless a bug specifically calls for it.\n","# Bug Bash\n\nAutonomously work through a project's bug\u002Fissue list, fixing bugs in priority order. Reproduce each bug, fix it, test it, commit it, and move on — without waiting for human input.\n\n## When to Use\n\n- Clearing a backlog of known bugs\n- Hardening before a release\n- Focused cleanup sprint\n\n## How to Invoke\n\n```\n\u002Fbug-bash                     # Fix all bugs, highest priority first\n\u002Fbug-bash P0-P1               # Only fix P0 and P1 (or critical\u002Fhigh) bugs\n\u002Fbug-bash --max 5             # Fix at most 5 bugs, then stop\n\u002Fbug-bash --dry-run           # Triage only — report what you'd fix, don't change code\n```\n\n---\n\n## Phase 0: Discover the Bug List\n\nFind the project's bug\u002Fissue tracker. Check these sources in order:\n\n1. **Local files**: Look for `bugs.md`, `BUGS.md`, `TODO.md`, `todo.md`, `ISSUES.md`, or similar in the repo root and `docs\u002F` directory\n2. **GitHub Issues**: Run `gh issue list --state open --limit 100` to check for a GitHub-based tracker\n3. **Project instructions**: Read `CLAUDE.md`, `CONTRIBUTING.md`, `README.md` for pointers to where bugs are tracked\n4. **Inline markers**: `grep -r \"TODO\\|FIXME\\|HACK\\|BUG\\|XXX\" --include=\"*.{py,js,ts,rb,go,rs,java,pm,pl}\" -l` as a last resort\n\nIf no bug list is found, tell the user and stop.\n\nOnce found, read the entire bug list. Parse out:\n- Each bug's **priority** (P0\u002FP1\u002FP2\u002FP3, critical\u002Fhigh\u002Fmedium\u002Flow, or whatever scheme the project uses)\n- Each bug's **status** (open, fixed, won't-fix, etc.)\n- Each bug's **description**\n\nFilter to open bugs only. Sort by priority (highest first), then by order of appearance. If the user specified a priority range or max count, apply those filters.\n\n---\n\n## Phase 1: Pre-flight Check\n\n**Work in your own worktree, never the shared checkout.** `git worktree add \u003Cpath> -b bug-bash\u002F\u003Cdate> main` and do everything there. Other agents may be working the same repo; the shared checkout is touched only to merge. Never `git stash` — `refs\u002Fstash` is shared across every worktree of a repo, and a sibling's `stash pop` can cross-apply into yours. And never edit another project's repository: if a bug's fix belongs in a dependency, file it with `\u002Frequest` and mark this one blocked.\n\nBefore touching any code, verify the project builds and tests pass in its current state.\n\n1. **Identify build\u002Ftest commands**: Read `Makefile`, `package.json`, `Cargo.toml`, `pyproject.toml`, `build.gradle`, etc. Look for the project's documented build and test commands (often in README, CONTRIBUTING, or CLAUDE.md).\n2. **Run the build**: Execute the build command. If it fails, STOP — tell the user the baseline is broken and list the failures.\n3. **Run the test suite**: Execute the test command. Record the results as the **baseline**.\n   - If tests fail, note which ones. These are **pre-existing failures** — do not count them against your fixes later.\n   - Record the exact test output for comparison after each fix.\n\nIf there is no build system or test suite, note this and proceed with extra caution — you'll rely on interactive testing and code review.\n\n---\n\n## Phase 2: Work Through Bugs\n\nFor each bug, in priority order:\n\n### Step 1: Understand\n\nEstablish expected versus actual behaviour and the code involved. Understand the current implementation before changing it.\n\n### Step 2: Reproduce\n\nBefore writing any fix, confirm the bug exists:\n- **Write a failing test** that demonstrates the bug, OR\n- **Reproduce interactively** if it's a UI\u002Fbehavioral issue (use tmux for TUI apps, curl for APIs, etc.), OR\n- **Read the code and confirm** the logical error if the bug is evident from inspection (e.g., wrong variable name, off-by-one)\n\n**Proof of bite.** A test that demonstrates the bug must actually fail against the unfixed code — run it *before* the fix and keep the failure output for the commit message. A guard that was never seen to fail is decoration: several have shipped that would have passed either way. If you prove it by tampering (reverting the fix, swapping in the old file), **commit the fix first** — an agent lost all its work to `git checkout --` mid-tamper — and the tampered build must compile and run, or you've proven nothing. Also write the case that must *not* trigger: a predicate tested only on the input its author wrote is unproven.\n\nIf you cannot reproduce the bug or confirm it exists in the current code:\n- It may already be fixed. Check recent commits.\n- It may be environment-specific. Note this.\n- Mark as \"Could not reproduce\" in the bug tracker and move on.\n\n### Step 3: Fix\n\nImplement the fix:\n- **One thesis per fix.** Fix the cause, not the symptom — a fix that only adds a guard should say in its commit body why the case was genuinely missing. Don't refactor the neighbourhood unless the fix requires it; if it does, the commit says so. Size isn't the measure; whether it shrinks or holds the learning surface is.\n- **Match existing patterns** for style, naming, and error handling.\n- **No drive-by changes.** If you spot another bug, add it to the tracker — don't fix it inline.\n\n### Step 4: Test\n\nVerify the fix thoroughly:\n\n1. **Confirm the failing test now passes** (or the interactive repro no longer reproduces)\n2. **Run the full test suite**: Compare against the baseline from Phase 1. If any previously-passing test now fails, your fix introduced a regression.\n   - If regression: **revert your changes**, note the conflict in the bug tracker (\"Fix attempted but caused regression in X — needs more careful approach\"), and move on.\n3. **Add or update tests** for the fixed behavior if a test didn't already exist\n4. **Drive the real thing** if the project has a recipe for it (`CLAUDE.md`'s verification recipe) — especially for UI bugs. Tests exercise the mechanism; this exercises the integration.\n\n### Step 5: Clean Up\n\nWhile the fix is fresh:\n- **Update documentation** if the bug or fix affects documented behavior\n- **Update the bug tracker** — mark the bug as fixed with a description of:\n  - What the root cause was\n  - What the fix was\n  - What tests were added\n  - Any decisions you made and why\n\n### Step 6: Commit\n\nCreate a commit for this bug fix:\n- Use a clear commit message referencing the bug (e.g., \"Fix P2: Shift+Tab cycles scope backward in find-in-files palette\")\n- Only include files related to this fix\n- Do NOT batch multiple bug fixes into one commit — one bug, one commit\n\n### Step 7: Next Bug\n\nMove to the next bug. Do not wait for human input between bugs. Keep going until:\n- All bugs in scope are processed, OR\n- The `--max` limit is reached, OR\n- You've hit a bug that is unsafe to fix without human guidance (see Phase 3)\n\n---\n\n## Phase 3: Skipping Bugs\n\nSkip a bug and move on if any of these apply:\n\n- **Ambiguous requirements**: The bug description is unclear and multiple interpretations lead to different fixes. Log: \"Skipped — ambiguous requirements. Needs clarification on [specific question].\"\n- **Too large**: The fix would require a major refactor or architectural change. Log: \"Skipped — fix requires [description of scope]. Not suitable for a bug bash.\"\n- **Conflicting requirements**: The fix would break another documented behavior or contradict another bug's resolution. Log: \"Skipped — conflicts with [other bug\u002Frequirement].\"\n- **Cannot reproduce**: The bug doesn't manifest in the current code or your environment. Log: \"Skipped — could not reproduce as of [commit hash].\"\n- **Dangerous**: The fix touches security-critical code, data migration, or deployment infrastructure in ways that need human review. Log: \"Skipped — needs human review for [reason].\"\n- **Blocked**: The fix depends on another bug being fixed first. Log: \"Skipped — blocked by [other bug]. Fix that one first.\"\n\nAlways log the reason in the bug tracker entry. Never silently skip.\n\n---\n\n## Phase 4: Session Summary\n\nAfter all bugs are processed (or the limit is reached), produce a summary:\n\n```\n# Bug Bash Summary\n\n**Date**: [date]\n**Scope**: [all \u002F P0-P1 \u002F max 5 \u002F etc.]\n**Baseline**: [commit hash at start]\n\n## Results\n\n| # | Priority | Bug | Result | Commit |\n|---|----------|-----|--------|--------|\n| 1 | P1 | Shell injection in Git.pm | FIXED | abc1234 |\n| 2 | P1 | Stale TODO.md references | FIXED | def5678 |\n| 3 | P2 | Shift+Tab broken in palette | FIXED | ghi9012 |\n| 4 | P2 | God object decomposition | SKIPPED | — |\n| 5 | P3 | Junk files not gitignored | FIXED | jkl3456 |\n\n**Fixed**: 4 | **Skipped**: 1 | **Regressed**: 0\n\n## Skipped Bugs\n- **P2: God object decomposition** — Too large. Fix requires architectural refactor across Editor.pm, Commands.pm, and Palette.pm. Not suitable for a bug bash.\n\n## Not Done\n- [Anything left out of a fix, deliberately or not: the second call site you didn't reach, the edge case you didn't cover, the test you couldn't make fail. An honest omissions list is worth more than the fixes — ask for it explicitly and it arrives; omit it and it never does.]\n\n## Decisions Made\n- [Any judgment calls you made during the bash, with reasoning]\n\n## Test Suite Status\n- Baseline: 927 tests, 926 passed, 1 pre-existing failure (syntax_samples.t)\n- Final: 934 tests, 933 passed, 1 pre-existing failure (syntax_samples.t)\n- New tests added: 7\n```\n\nPush the branch if on a feature branch. Report the summary to the user.\n\n---\n\n## Decision-Making Guidelines\n\nYou are working unassisted. When a decision is required:\n\n1. **Prefer the option you can explain.** If unsure between two approaches, pick the one whose thesis fits in a sentence and doesn't grow the public surface — not simply the one with the smaller diff.\n2. **Follow existing patterns.** If the codebase does X one way everywhere, do it the same way — even if you think another way is better.\n3. **Don't gold-plate.** Fix the bug as described. Don't add features, extra configurability, or \"while I'm here\" improvements.\n4. **When in doubt, skip.** It's better to skip a bug and let a human decide than to make a bad fix that introduces a regression.\n5. **Log everything.** Every decision, every skip reason, every judgment call goes in the bug tracker and commit message. A human will review your work later.\n\n---\n\n## Important Safety Rules\n\n- **Never force-push.** Always use regular `git push`.\n- **Never `git stash`, never edit another project's checkout.** See Phase 1.\n- **Never modify git config.** Don't change user.name, user.email, or any git settings.\n- **One commit per bug.** Don't squash or amend across bugs.\n- **Revert on regression.** If your fix breaks something, revert it — don't try to fix the fix.\n- **Respect pre-commit hooks.** If a hook fails, fix the issue and create a new commit (don't `--no-verify`).\n- **Don't delete test files.** Even if they seem broken or outdated.\n- **Don't touch CI\u002FCD config** unless a bug specifically calls for it.\n",{"name":5,"description":7,"argument-hint":35},[36],"P0-P1 | --max N | --dry-run","main","0d32bda2cb8e014372c2814f5aa47f48eee0f411","skills\u002Fbug-bash","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills","https:\u002F\u002Fgithub.com\u002Fjoewalnes","Experimental Projects @ Stripe.\r\n\r\nPreviously: Apple, Google, ThoughtWorks, DRW, Startups.","2026-09-09 00:56:55",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":56},[46,49,52],{"tool":47,"command":48},"npx","npx skills add joewalnes\u002Fskills --skill bug-bash",{"tool":50,"command":51},"bunx","bunx skills add joewalnes\u002Fskills --skill bug-bash",{"tool":53,"command":54},"pnpm","pnpm dlx skills add joewalnes\u002Fskills --skill bug-bash","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fbug-bash\u002FSKILL.md",true,[],[59,75,85,104,118,124],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":30},"01M6K7MSG99DXYKTVVFGZWND1S","tool-web","Build a lightweight single-file web application with no external dependencies","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftool-web\u002FSKILL.md",[],[66,67,68,69,70,71,72,73,74],"tool web","tool","web","file","single-file","build","dependencies","application","with",{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":30},"01M6K7MPN2FG6M6JQFWFR18CJ0","todo","Add a new bug or todo to the project's tracker","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftodo\u002FSKILL.md",[],[77,20,82,83,26,84],"tracker","task","add",{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":30},"01M6K7MGQFFYCD8HFDP1B95KDN","slop","Find unchosen code — surface area growing without leverage. Use pre-merge on a branch, or to check if a repo is accreting.","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fslop\u002FSKILL.md",[],[87,92,93,94,95,96,97,98,99,100,101,102,103],"pre-merge","rev","explain","merge","code","find","branch","repo","check","use","leverage","quick",{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":30},"01M6K7MCVREA4NZ46Q9JWZJ7TG","sitrep","Where were we? Branch, in-progress work, blocked agents, inbox, next steps. Use when resuming a session.","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fsitrep\u002FSKILL.md",[],[106,23,111,19,112,113,101,114,115,116,98,117],"grep","inbox","session","next","progress","next steps","read",{"skillNo":119,"name":120,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":30},"01M6K7M99FYMH9SFSEKB718J2B","scorecard","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fscorecard\u002FSKILL.md",[],[120],{"skillNo":125,"name":126,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":129,"publishedAt":30},"01M6K7M4HNZR34GSB3V14KVCWA","retro","https:\u002F\u002Fgithub.com\u002Fjoewalnes\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fretro\u002FSKILL.md",[],[126]]