[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79D67XKJ2EX8SVMWC798TDA":3,"skill-same:01M79D67XKJ2EX8SVMWC798TDA":48},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":14,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M79D67XKJ2EX8SVMWC798TDA","check-work",null,"Check your work with a verification subagent that reviews diffs, runs builds and tests, and evaluates correctness. Read this file for instructions. Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"\u002Fcheck-work\", \"\u002Fcheck\", \"\u002Fverify\", or \"\u002Fself-verify\".","zfifteen\u002Fskills-catalog","zfifteen","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F221906715?v=4","01M79CQFXDZM5J5CJTCECPFA66","Dionisio Alberto Lopez III","01M79CQFY4C62PN8A2T5KDAHGA",1,"MIT","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fcheck-work\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: check-work\ndescription: >\n  Check your work with a verification subagent that reviews diffs, runs builds\n  and tests, and evaluates correctness. Read this file for instructions. Use when\n  asked to \"check work\", \"verify changes\", \"self-verify\", \"\u002Fcheck-work\", \"\u002Fcheck\",\n  \"\u002Fverify\", or \"\u002Fself-verify\".\nmetadata:\n  short-description: \"Verify changes with a subagent\"\n---\n\n# \u002Fcheck-work -- Self-Verification\n\nVerify work by spawning a verifier subagent, checking its verdict, and\nfixing issues until it passes.\n\n## Usage\n\n`\u002Fcheck-work [focus area]`\n\nThe optional focus area tells the verifier to pay special attention to specific\naspects of the changes (e.g. \"auth logic and JWT handling\").\n\n## Mode Detection\n\nDetermine which mode you are in before proceeding:\n\n- **Same-turn mode**: There is a user task alongside this skill (e.g. headless\n  `--check`). **Complete the task fully first**, then proceed to Step 1 below.\n- **Standalone mode**: There is no task — just `\u002Fcheck-work` (or the alias `\u002Fcheck`) or the skill was invoked\n  after a previous turn. Proceed directly to Step 1.\n\n## Steps\n\n1. Call the `task` tool with:\n   - `description`: must start with `\"[checking my work]\"` followed by a short label\n   - `subagent_type`: `\"general-purpose\"`\n   - `run_in_background`: `false`\n   - `prompt`: copy the **VERIFIER PROMPT** section below verbatim. If a focus\n     area was specified by the user, append this to the prompt:\n     ```\n     ## Additional Focus\n     \u003Cfocus area text>\n     Pay special attention to these areas during verification.\n     ```\n\n2. Read the subagent's result. Look for `VERDICT: PASS` or `VERDICT: FAIL`.\n\n3. If **PASS**: summarize what the verifier confirmed and stop.\n\n4. If **FAIL** (or no verdict found): fix the issues the verifier identified,\n   then go back to step 1. Repeat up to 3 times.\n\n## VERIFIER PROMPT\n\nYou are an expert verifier. Your job is to determine whether the work done in\nthis session correctly and completely addresses the user's requests.\n\nYou already have the full conversation context, so you know what the user asked\nfor, what approach was taken, what tools were used, and what outcomes were\nobserved. You also have full access to the same environment and tools the\noriginal agent had.\n\n=== SCOPE ===\n\nDetermine what to verify:\n\n- If a **focus area** was specified (see Additional Focus below), verify that\n  specific area. Use the full session trace for context -- understand what was\n  asked, what was done, and what state the environment is in -- but scope your\n  verdict to the focused area.\n- If no focus area was specified, verify **all work done in this session**.\n\n=== WORKFLOW ===\n\nEvery verification runs two phases. Phase A (Trace Review) always runs.\nPhase B (Code Review) runs when code review is relevant to the task.\n\n--- PHASE A: TRACE REVIEW ---\n\nThis phase reviews what the agent did, whether it completed all tasks, and\nwhether its outputs were correct. Run this for every verification.\n\n1. UNDERSTAND THE REQUEST:\n   Read through the conversation to identify everything the user asked for --\n   not just the first message, but follow-up requests, corrections, and\n   clarifications across the entire session. Restate these as a concrete\n   checklist of deliverables or success criteria.\n\n   Include all task types:\n   - Code tasks (implement feature, fix bug, refactor)\n   - Operational tasks (submit the eval job, deploy to staging, kick off CI)\n   - Git\u002FPR tasks (push the branch, create the PR, address review comments)\n   - Research tasks (analyze data, investigate a failure, find root cause)\n   - Q&A tasks (explain how X works, compare approaches, answer a question)\n   - Configuration tasks (update settings, add environment variables, modify configs)\n\n   If a focus area was specified, the checklist should center on that area\n   but include related items that affect the verdict.\n\n2. RECONSTRUCT WHAT HAPPENED:\n   Trace the actions the agent actually took. For each tool call, command, or\n   action in the conversation, identify what the outcome was. Look for:\n   - Actions that failed or produced unexpected results\n   - Things the user asked for that were never attempted\n   - Things the agent said it would do but did not actually do\n   - Work the agent deferred to the user that it could have done itself\n     (e.g. printing instructions instead of running a command)\n   - Questions answered incorrectly or incompletely\n   - Reasoning errors in the agent's analysis or explanations\n\n3. VERIFY CURRENT STATE:\n   Gather evidence about what actually happened by inspecting the environment\n   yourself. Do not trust the conversation's claims -- verify them:\n   - If the session involved code changes, read the modified files.\n   - If the session involved submitting jobs or API calls, check their status.\n   - If the session involved running commands, verify their effects.\n   - If the session involved creating resources (PRs, branches, configs),\n     confirm they exist and are in the expected state.\n   - If the session involved answering questions, verify the answers are\n     correct by checking the source material yourself.\n\n--- PHASE B: CODE REVIEW ---\n\nRun this phase when the task involves code in any way. Examples:\n- The agent wrote or modified code during this session\n- The user asked the agent to review existing code (security audit,\n  code review, architecture review)\n- The task involved evaluating code correctness, performance, or security\n- The changes include code-like configuration (BUILD files, CI configs,\n  k8s manifests, IaC)\n\nSkip this phase only if the session was purely non-code with no code\ninvolvement at all (general Q&A, operational tasks with no code context,\ndata analysis, research).\n\n4. COLLECT THE DIFF OR READ THE CODE:\n   If code was written or modified: run `git diff` to see unstaged changes.\n   Run `git diff --cached` to see staged changes. Run `git log --oneline -3`\n   and `git diff HEAD~1..HEAD` to check for recent commits. Combine these to\n   get the full picture of all changes made during this session.\n\n   If the session was a code review of existing code (no modifications): read\n   the files the agent reviewed. You need the actual source to verify whether\n   the agent's analysis was correct and thorough.\n\n   In both cases, read the relevant files and their surrounding context to\n   understand the scope.\n\n5. EVALUATE THE CODE:\n   Consider the following criteria carefully:\n\n   a) CORRECTNESS: If code was written or modified -- does it compile, run,\n      and pass tests? A broken build or failing tests is an automatic FAIL.\n      If this was a review of existing code -- was the agent's assessment of\n      correctness accurate?\n\n   b) ADEQUACY: Do the changes or the review adequately address the user's\n      request? Are all requested features implemented, fixes applied, or\n      review areas covered? Were all non-code tasks completed (not just the\n      code part)? There could be several possible correct solutions -- all\n      correct solutions should be considered valid.\n\n   c) EXCESS: Do the changes do anything in excess that could negatively\n      impact the codebase? Unnecessary refactors, added complexity, unrelated\n      modifications, or gold-plating beyond what was asked.\n\n   d) EDGE CASES: Do the changes sufficiently handle edge cases without being\n      overly verbose or complex? Missing critical edge cases is a problem, but\n      over-engineering for hypothetical scenarios is also a problem.\n\n6. BUILD AND TEST:\n   Read the repo's AGENTS.md \u002F Claude.md (the root file and any in the\n   directories of changed files) and README for build\u002Ftest commands. Run them:\n   - Build the project (e.g. cargo check, npm run build, tsc). A broken build\n     is an automatic FAIL.\n   - Run the test suite (e.g. cargo test, pytest, npm test). Failing tests are\n     an automatic FAIL.\n   - Run linters\u002Ftype-checkers if configured (cargo clippy, eslint, mypy, tsc).\n\n7. DESIGN AND RUN VERIFICATION CHECKS:\n   You are encouraged to write and run your own tests or checks to verify the\n   work is correct. This may include:\n   - Writing small test scripts that exercise new\u002Fchanged functionality\n   - Running the application and exercising it (curl endpoints, invoke CLIs)\n   - Adding assertions that confirm the expected behavior\n   - Checking boundary conditions and error paths\n   - Querying APIs or services to confirm actions were completed\n\n   You may need to run several tool calls, tests, checks, or other analysis\n   to determine correctness. Take your time -- thoroughness matters more\n   than speed.\n\n8. REVIEW THE CODE:\n   Read the diff (or the reviewed files) and surrounding source for context.\n   If code was written, look for issues the agent introduced. If the agent\n   reviewed existing code, verify the agent's findings are correct and check\n   for issues the agent missed. In both cases look for:\n   - Bugs: logic errors, off-by-one, null\u002Fundefined access, unhandled errors\n   - Security: injection, XSS, unsafe deserialization, secrets in code\n   - Missing validation at system boundaries (user input, API responses)\n   - Regressions: did the change break existing behavior?\n   - Test quality: are new tests circular, over-mocked, or only covering\n     happy paths?\n   - Project-instruction compliance: where the repo's AGENTS.md \u002F Claude.md\n     files (read in step 6) state reviewable rules (style, structure, naming,\n     conventions, policy), a change that violates one is a FAIL -- cite the\n     rule and file:line. If they state no review-relevant rules, do not invent\n     violations.\n\n--- VERDICT ---\n\n9. VERDICT:\n   After completing your analysis, end your response with exactly one of:\n   VERDICT: PASS -- the work correctly and adequately addresses the user's requests\n   VERDICT: FAIL -- there are issues that need fixing\n\n   If FAIL, describe what is broken, the exact error output, and what\n   specifically needs to change. Be precise about file paths and line numbers\n   for code issues, and specific about what was missed or incorrect for\n   non-code issues.\n\n   If PASS, describe the verification process and what evidence confirms\n   success.\n\n=== IMPORTANT PRINCIPLES ===\n\n- Think through problems step by step. When you are unsure, gather more\n  information before concluding.\n- You should assume that if the code fails to compile or run, the changes do\n  not address the user's request.\n- Verify outcomes, not just code. If the user asked \"submit the eval job\",\n  check whether the job was actually submitted and accepted -- do not just\n  verify that the code change that enables submission is correct.\n- Do not accept proxy signals as proof of completion. Passing tests, a\n  successful build, or substantial effort are useful evidence only if they\n  cover every requirement in the checklist.\n- Do not invent issues to fill space. If the work genuinely addresses the\n  user's requests correctly, say PASS. Nitpicks about style or theoretical\n  concerns that do not affect correctness should not cause a FAIL. However,\n  violations of rules explicitly stated in the repo's AGENTS.md \u002F Claude.md\n  are policy, not nitpicks, and DO cause a FAIL.\n- Focus on whether the work addresses what the user actually asked for, not\n  on what you might have done differently.\n- Any temporary test files or modifications you create for verification\n  purposes are fine -- they will not affect the parent agent's workspace.\n\n=== OUTPUT FORMAT ===\n\nWrite a structured verification report:\n\n## Checklist\nThe user's requirements restated as a numbered list of concrete items.\nInclude all task types (code, operational, research, Q&A, etc.).\n\n## Action Trace\nFor each checklist item: what was done, what tools\u002Fcommands were used, and\nwhether the action succeeded. Note any items that were not attempted, answered\nincorrectly, or deferred to the user.\n\n## Diff Summary \u002F Code Scope (Phase B only)\nIf code was written: brief description of what files changed and the scope.\nIf code was reviewed: which files were reviewed and what areas were covered.\n\n## Evaluation\nAssessment against each applicable criterion:\n- **Correctness**: Does it compile, run, pass tests? (Phase B)\n- **Adequacy**: Does it address the user's request? Were all tasks completed?\n- **Excess**: Any unnecessary changes? (Phase B)\n- **Edge Cases**: Sufficient coverage without over-engineering? (Phase B)\n\n## Build & Test Results (Phase B only)\nOutput from builds, tests, and linters. Include exact command and result.\n\n## Issues\nFor each issue found (skip this section entirely if none):\n\n### Issue N -- Severity: bug\u002Fgap\u002Fregression\u002Fsuggestion\n- File: path\u002Fto\u002Ffile.ext:LINE (for code issues)\n- Description: what is wrong\n- Evidence: exact error output, missing action, or incorrect answer\n- Suggestion: how to fix\n\nThen end with exactly:\nVERDICT: PASS\nor\nVERDICT: FAIL\n","# \u002Fcheck-work -- Self-Verification\n\nVerify work by spawning a verifier subagent, checking its verdict, and\nfixing issues until it passes.\n\n## Usage\n\n`\u002Fcheck-work [focus area]`\n\nThe optional focus area tells the verifier to pay special attention to specific\naspects of the changes (e.g. \"auth logic and JWT handling\").\n\n## Mode Detection\n\nDetermine which mode you are in before proceeding:\n\n- **Same-turn mode**: There is a user task alongside this skill (e.g. headless\n  `--check`). **Complete the task fully first**, then proceed to Step 1 below.\n- **Standalone mode**: There is no task — just `\u002Fcheck-work` (or the alias `\u002Fcheck`) or the skill was invoked\n  after a previous turn. Proceed directly to Step 1.\n\n## Steps\n\n1. Call the `task` tool with:\n   - `description`: must start with `\"[checking my work]\"` followed by a short label\n   - `subagent_type`: `\"general-purpose\"`\n   - `run_in_background`: `false`\n   - `prompt`: copy the **VERIFIER PROMPT** section below verbatim. If a focus\n     area was specified by the user, append this to the prompt:\n     ```\n     ## Additional Focus\n     \u003Cfocus area text>\n     Pay special attention to these areas during verification.\n     ```\n\n2. Read the subagent's result. Look for `VERDICT: PASS` or `VERDICT: FAIL`.\n\n3. If **PASS**: summarize what the verifier confirmed and stop.\n\n4. If **FAIL** (or no verdict found): fix the issues the verifier identified,\n   then go back to step 1. Repeat up to 3 times.\n\n## VERIFIER PROMPT\n\nYou are an expert verifier. Your job is to determine whether the work done in\nthis session correctly and completely addresses the user's requests.\n\nYou already have the full conversation context, so you know what the user asked\nfor, what approach was taken, what tools were used, and what outcomes were\nobserved. You also have full access to the same environment and tools the\noriginal agent had.\n\n=== SCOPE ===\n\nDetermine what to verify:\n\n- If a **focus area** was specified (see Additional Focus below), verify that\n  specific area. Use the full session trace for context -- understand what was\n  asked, what was done, and what state the environment is in -- but scope your\n  verdict to the focused area.\n- If no focus area was specified, verify **all work done in this session**.\n\n=== WORKFLOW ===\n\nEvery verification runs two phases. Phase A (Trace Review) always runs.\nPhase B (Code Review) runs when code review is relevant to the task.\n\n--- PHASE A: TRACE REVIEW ---\n\nThis phase reviews what the agent did, whether it completed all tasks, and\nwhether its outputs were correct. Run this for every verification.\n\n1. UNDERSTAND THE REQUEST:\n   Read through the conversation to identify everything the user asked for --\n   not just the first message, but follow-up requests, corrections, and\n   clarifications across the entire session. Restate these as a concrete\n   checklist of deliverables or success criteria.\n\n   Include all task types:\n   - Code tasks (implement feature, fix bug, refactor)\n   - Operational tasks (submit the eval job, deploy to staging, kick off CI)\n   - Git\u002FPR tasks (push the branch, create the PR, address review comments)\n   - Research tasks (analyze data, investigate a failure, find root cause)\n   - Q&A tasks (explain how X works, compare approaches, answer a question)\n   - Configuration tasks (update settings, add environment variables, modify configs)\n\n   If a focus area was specified, the checklist should center on that area\n   but include related items that affect the verdict.\n\n2. RECONSTRUCT WHAT HAPPENED:\n   Trace the actions the agent actually took. For each tool call, command, or\n   action in the conversation, identify what the outcome was. Look for:\n   - Actions that failed or produced unexpected results\n   - Things the user asked for that were never attempted\n   - Things the agent said it would do but did not actually do\n   - Work the agent deferred to the user that it could have done itself\n     (e.g. printing instructions instead of running a command)\n   - Questions answered incorrectly or incompletely\n   - Reasoning errors in the agent's analysis or explanations\n\n3. VERIFY CURRENT STATE:\n   Gather evidence about what actually happened by inspecting the environment\n   yourself. Do not trust the conversation's claims -- verify them:\n   - If the session involved code changes, read the modified files.\n   - If the session involved submitting jobs or API calls, check their status.\n   - If the session involved running commands, verify their effects.\n   - If the session involved creating resources (PRs, branches, configs),\n     confirm they exist and are in the expected state.\n   - If the session involved answering questions, verify the answers are\n     correct by checking the source material yourself.\n\n--- PHASE B: CODE REVIEW ---\n\nRun this phase when the task involves code in any way. Examples:\n- The agent wrote or modified code during this session\n- The user asked the agent to review existing code (security audit,\n  code review, architecture review)\n- The task involved evaluating code correctness, performance, or security\n- The changes include code-like configuration (BUILD files, CI configs,\n  k8s manifests, IaC)\n\nSkip this phase only if the session was purely non-code with no code\ninvolvement at all (general Q&A, operational tasks with no code context,\ndata analysis, research).\n\n4. COLLECT THE DIFF OR READ THE CODE:\n   If code was written or modified: run `git diff` to see unstaged changes.\n   Run `git diff --cached` to see staged changes. Run `git log --oneline -3`\n   and `git diff HEAD~1..HEAD` to check for recent commits. Combine these to\n   get the full picture of all changes made during this session.\n\n   If the session was a code review of existing code (no modifications): read\n   the files the agent reviewed. You need the actual source to verify whether\n   the agent's analysis was correct and thorough.\n\n   In both cases, read the relevant files and their surrounding context to\n   understand the scope.\n\n5. EVALUATE THE CODE:\n   Consider the following criteria carefully:\n\n   a) CORRECTNESS: If code was written or modified -- does it compile, run,\n      and pass tests? A broken build or failing tests is an automatic FAIL.\n      If this was a review of existing code -- was the agent's assessment of\n      correctness accurate?\n\n   b) ADEQUACY: Do the changes or the review adequately address the user's\n      request? Are all requested features implemented, fixes applied, or\n      review areas covered? Were all non-code tasks completed (not just the\n      code part)? There could be several possible correct solutions -- all\n      correct solutions should be considered valid.\n\n   c) EXCESS: Do the changes do anything in excess that could negatively\n      impact the codebase? Unnecessary refactors, added complexity, unrelated\n      modifications, or gold-plating beyond what was asked.\n\n   d) EDGE CASES: Do the changes sufficiently handle edge cases without being\n      overly verbose or complex? Missing critical edge cases is a problem, but\n      over-engineering for hypothetical scenarios is also a problem.\n\n6. BUILD AND TEST:\n   Read the repo's AGENTS.md \u002F Claude.md (the root file and any in the\n   directories of changed files) and README for build\u002Ftest commands. Run them:\n   - Build the project (e.g. cargo check, npm run build, tsc). A broken build\n     is an automatic FAIL.\n   - Run the test suite (e.g. cargo test, pytest, npm test). Failing tests are\n     an automatic FAIL.\n   - Run linters\u002Ftype-checkers if configured (cargo clippy, eslint, mypy, tsc).\n\n7. DESIGN AND RUN VERIFICATION CHECKS:\n   You are encouraged to write and run your own tests or checks to verify the\n   work is correct. This may include:\n   - Writing small test scripts that exercise new\u002Fchanged functionality\n   - Running the application and exercising it (curl endpoints, invoke CLIs)\n   - Adding assertions that confirm the expected behavior\n   - Checking boundary conditions and error paths\n   - Querying APIs or services to confirm actions were completed\n\n   You may need to run several tool calls, tests, checks, or other analysis\n   to determine correctness. Take your time -- thoroughness matters more\n   than speed.\n\n8. REVIEW THE CODE:\n   Read the diff (or the reviewed files) and surrounding source for context.\n   If code was written, look for issues the agent introduced. If the agent\n   reviewed existing code, verify the agent's findings are correct and check\n   for issues the agent missed. In both cases look for:\n   - Bugs: logic errors, off-by-one, null\u002Fundefined access, unhandled errors\n   - Security: injection, XSS, unsafe deserialization, secrets in code\n   - Missing validation at system boundaries (user input, API responses)\n   - Regressions: did the change break existing behavior?\n   - Test quality: are new tests circular, over-mocked, or only covering\n     happy paths?\n   - Project-instruction compliance: where the repo's AGENTS.md \u002F Claude.md\n     files (read in step 6) state reviewable rules (style, structure, naming,\n     conventions, policy), a change that violates one is a FAIL -- cite the\n     rule and file:line. If they state no review-relevant rules, do not invent\n     violations.\n\n--- VERDICT ---\n\n9. VERDICT:\n   After completing your analysis, end your response with exactly one of:\n   VERDICT: PASS -- the work correctly and adequately addresses the user's requests\n   VERDICT: FAIL -- there are issues that need fixing\n\n   If FAIL, describe what is broken, the exact error output, and what\n   specifically needs to change. Be precise about file paths and line numbers\n   for code issues, and specific about what was missed or incorrect for\n   non-code issues.\n\n   If PASS, describe the verification process and what evidence confirms\n   success.\n\n=== IMPORTANT PRINCIPLES ===\n\n- Think through problems step by step. When you are unsure, gather more\n  information before concluding.\n- You should assume that if the code fails to compile or run, the changes do\n  not address the user's request.\n- Verify outcomes, not just code. If the user asked \"submit the eval job\",\n  check whether the job was actually submitted and accepted -- do not just\n  verify that the code change that enables submission is correct.\n- Do not accept proxy signals as proof of completion. Passing tests, a\n  successful build, or substantial effort are useful evidence only if they\n  cover every requirement in the checklist.\n- Do not invent issues to fill space. If the work genuinely addresses the\n  user's requests correctly, say PASS. Nitpicks about style or theoretical\n  concerns that do not affect correctness should not cause a FAIL. However,\n  violations of rules explicitly stated in the repo's AGENTS.md \u002F Claude.md\n  are policy, not nitpicks, and DO cause a FAIL.\n- Focus on whether the work addresses what the user actually asked for, not\n  on what you might have done differently.\n- Any temporary test files or modifications you create for verification\n  purposes are fine -- they will not affect the parent agent's workspace.\n\n=== OUTPUT FORMAT ===\n\nWrite a structured verification report:\n\n## Checklist\nThe user's requirements restated as a numbered list of concrete items.\nInclude all task types (code, operational, research, Q&A, etc.).\n\n## Action Trace\nFor each checklist item: what was done, what tools\u002Fcommands were used, and\nwhether the action succeeded. Note any items that were not attempted, answered\nincorrectly, or deferred to the user.\n\n## Diff Summary \u002F Code Scope (Phase B only)\nIf code was written: brief description of what files changed and the scope.\nIf code was reviewed: which files were reviewed and what areas were covered.\n\n## Evaluation\nAssessment against each applicable criterion:\n- **Correctness**: Does it compile, run, pass tests? (Phase B)\n- **Adequacy**: Does it address the user's request? Were all tasks completed?\n- **Excess**: Any unnecessary changes? (Phase B)\n- **Edge Cases**: Sufficient coverage without over-engineering? (Phase B)\n\n## Build & Test Results (Phase B only)\nOutput from builds, tests, and linters. Include exact command and result.\n\n## Issues\nFor each issue found (skip this section entirely if none):\n\n### Issue N -- Severity: bug\u002Fgap\u002Fregression\u002Fsuggestion\n- File: path\u002Fto\u002Ffile.ext:LINE (for code issues)\n- Description: what is wrong\n- Evidence: exact error output, missing action, or incorrect answer\n- Suggestion: how to fix\n\nThen end with exactly:\nVERDICT: PASS\nor\nVERDICT: FAIL\n",{"name":5,"metadata":24,"description":26},{"short-description":25},"Verify changes with a subagent","Check your work with a verification subagent that reviews diffs, runs builds and tests, and evaluates correctness. Read this file for instructions. Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"\u002Fcheck-work\", \"\u002Fcheck\", \"\u002Fverify\", or \"\u002Fself-verify\".\n","main","d9535210da9af28aa76bca80de80355197e1b6d6","skills\u002Fcheck-work","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog","https:\u002F\u002Fgithub.com\u002Fzfifteen","I like to measure things.","2026-07-19 23:41:54",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add zfifteen\u002Fskills-catalog --skill check-work",{"tool":40,"command":41},"bunx","bunx skills add zfifteen\u002Fskills-catalog --skill check-work",{"tool":43,"command":44},"pnpm","pnpm dlx skills add zfifteen\u002Fskills-catalog --skill check-work","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fcheck-work\u002FSKILL.md",true,[],[49,56,63,70,79,86],{"skillNo":50,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":53,"categories":54,"keywords":55,"publishedAt":19},"01M79GXQ2PWPT2E4CJYHXVJ0YE","Zotero","Use Zotero Desktop from Codex to enable\u002Fprobe the local API, search a local Zotero library, list items\u002Fcollections\u002Ftags, export BibTeX, insert citation keys into LaTeX or Markdown drafts, read indexed full text when requested, and import BibTeX\u002FRIS records into Zotero through the connector server. Use when the user mentions Zotero, citations, references.bib, BibTeX export, local Zotero API, localhost:23119, or adding citations from a Zotero library.","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fzotero\u002FSKILL.md",[],[],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M79GXN85NETKBE06865CWK6Z","zoom-apps-sdk","Use when using Apps SDK.","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fzoom-apps-sdk\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M79GXCYD7FGYS932J9N9RFFY","youtube","Obtain accurate, timestamped YouTube video transcripts (preferring manual captions over auto-generated) as the primary deliverable, then derive summaries, key claims, or notes from that transcript.  Use when the user provides a YouTube URL (or video ID) and wants the transcript, a faithful summary, key claims with attribution, timestamped notes, or \"what does this video actually say\".  The skill is transcript-first: never substitute metadata, title, description, or search snippets for the spoken content.  Triggers: \"youtube transcript\", \"get the transcript of this video\", \"summarize this youtube video\", \"what did they say in https:\u002F\u002Fyoutube.com\u002F...\", \"key claims from this talk\".","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":74,"licenseFileUrl":75,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M79GX8EXHHYM566BKR7MS778","yeet","Publish local changes to GitHub by confirming scope, committing intentionally, pushing the branch, and opening a draft PR through the GitHub app from this plugin, with `gh` used only as a fallback where connector coverage is insufficient.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fyeet\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fyeet\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M79GX2HFJPX84NMRE5VBTYX9","x-post-creator","Create X-ready post threads and co-located story packages from the active conversation or referenced project material. Use when the user says turn this into an X-post, \u002Fx-post, make this into an X thread, or wants plain-language social posts with supporting images saved to docs\u002Fx-posts\u002F[story-name]\u002F.","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fx-post-creator\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M79GWXQT4HBT6QME6QSBZJFA","x-api","Search, read, and write X posts through the local X API MCP server (the same backend used by the Codex x-api skill). Use for searching X\u002Fx.com, reading threads and replies, creating lists, staging\u002Fpublishing posts for @alltheputs, finding high-visibility reply targets for research outreach, ingesting public context, or any request that mentions \u002Fx-api or \"search X\".","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fx-api\u002FSKILL.md",[],[]]