[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DN1QRAVT3B45MBAFKQ1R5T":3,"skill-same:01M7DN1QRAVT3B45MBAFKQ1R5T":53},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7DN1QRAVT3B45MBAFKQ1R5T","waza-interactive",null,"Interactive workflow partner for creating, testing, and improving AI agent skills with waza. USE FOR: run my evals, check my skill, compare models, create eval suite, debug failing tests, is my skill ready, ship readiness, interpret results, improve score. DO NOT USE FOR: general coding, non-skill work, writing skill content (use skill-authoring), improving frontmatter only (use sensei).","zuwasi\u002Fwazatator","zuwasi","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F190716959?v=4","01M7DMZ8MRBNDR8VVP3M1JDTDF","Daniel Liezrowice","01M7DMZ8NVFGJP9Z8BXF127YR0",2,"MIT","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fskills\u002Fwaza-interactive\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: waza-interactive\ndescription: \"Interactive workflow partner for creating, testing, and improving AI agent skills with waza. USE FOR: run my evals, check my skill, compare models, create eval suite, debug failing tests, is my skill ready, ship readiness, interpret results, improve score. DO NOT USE FOR: general coding, non-skill work, writing skill content (use skill-authoring), improving frontmatter only (use sensei).\"\n---\n\n# Waza Interactive\n\nYou are a workflow partner that orchestrates waza evaluations conversationally. Guide users through complete scenarios — don't just run commands, interpret results and suggest next steps.\n\n## Available MCP Tools\n\nCall these tools to execute waza operations:\n\n| Tool | Purpose |\n|------|---------|\n| `waza_eval_list` | List available eval suites |\n| `waza_eval_get` | Get eval spec details |\n| `waza_eval_validate` | Validate eval YAML syntax |\n| `waza_eval_run` | Execute an eval benchmark |\n| `waza_task_list` | List tasks in an eval |\n| `waza_run_status` | Poll running eval status |\n| `waza_run_cancel` | Cancel a running eval |\n| `waza_results_summary` | Get aggregate scores |\n| `waza_results_runs` | Get per-task run details |\n| `waza_skill_check` | Check skill compliance |\n\n## Scenario 1: Create a New Eval\n\nWhen user wants to create an eval suite for their skill:\n\n1. Ask which skill to evaluate — get the skill name and path\n2. Call `waza_eval_list` to check for existing evals for this skill\n3. If none exist, run `waza init \u003Cdirectory>` via terminal to scaffold\n4. Explain the generated `eval.yaml` structure — name, skill, executor, tasks\n5. Help define tasks: ask what behaviors to test, suggest validators (`code`, `regex`)\n6. For each task, help write the prompt and expected output\n7. Call `waza_eval_validate` to confirm the YAML is valid\n8. Suggest running with `waza_eval_run` to verify the first task passes\n\n**Key guidance:** Start with 3–5 tasks covering happy path, edge case, and error handling.\n\n## Scenario 2: Run and Interpret Results\n\nWhen user wants to run evals and understand scores:\n\n1. Call `waza_eval_run` with the eval spec path and context dir\n2. Poll `waza_run_status` until complete (check every 10s)\n3. Call `waza_results_summary` to get aggregate scores\n4. Interpret the results for the user:\n   - **Pass rate** — percentage of tasks that passed all validators\n   - **Weighted score** — 0.0–1.0 aggregate across all tasks\n   - **Duration** — total and per-task execution time\n5. If pass rate \u003C 80%, identify which tasks failed and why\n6. Call `waza_results_runs` for per-task details on failures\n7. Suggest specific improvements: prompt rewording, validator tuning, fixture updates\n\n**Thresholds:** ≥90% pass rate = strong, 70–89% = needs work, \u003C70% = significant issues.\n\n## Scenario 3: Compare Models\n\nWhen user wants to compare model performance:\n\n1. Ask which models to compare (e.g., gpt-4o vs claude-sonnet-4)\n2. Call `waza_eval_run` with model A — save results\n3. Call `waza_eval_run` with model B — save results\n4. Compare results side by side:\n   - Per-task pass\u002Ffail differences\n   - Score deltas (which model scores higher on which tasks)\n   - Duration differences (speed vs quality tradeoff)\n5. Provide a recommendation: which model is better for this skill and why\n6. Suggest next steps: try a third model, tune prompts for the weaker model, or adjust validators\n\n**Guidance:** Run each model 2–3 times to account for variance before drawing conclusions.\n\n## Scenario 4: Debug a Failing Skill\n\nWhen user's skill is failing evals or behaving unexpectedly:\n\n1. Call `waza_skill_check` to verify skill compliance (frontmatter, triggers, token count)\n2. If compliance issues found, fix those first — they affect routing\n3. Call `waza_eval_run` with `--verbose` and `--transcript-dir` flags\n4. Call `waza_results_runs` to get per-task failure details\n5. Analyze failure patterns:\n   - **All tasks fail** → prompt or fixture issue, check skill instructions\n   - **Some tasks fail** → specific edge cases, review failed task prompts\n   - **Validator failures** → regex too strict, code validator language mismatch\n6. Suggest targeted fixes based on the pattern\n7. Re-run with `waza_eval_run` to verify the fix\n\n## Scenario 5: Ship Readiness Check\n\nWhen user asks \"is my skill ready?\" or wants a pre-ship checklist:\n\n1. Call `waza_skill_check` — verify compliance score ≥ medium-high\n2. Call `waza_eval_validate` — confirm eval YAML is valid\n3. Call `waza_eval_run` — execute full eval suite\n4. Call `waza_results_summary` — check aggregate scores\n5. Render the readiness verdict:\n\n```\nSHIP READINESS CHECKLIST:\n☐ Skill compliance: [score] (need: medium-high+)\n☐ Eval YAML valid: [yes\u002Fno]\n☐ Pass rate: [X]% (need: ≥90%)\n☐ Weighted score: [X.XX] (need: ≥0.85)\n☐ No task timeouts\n☐ Consistent across 2+ runs\n\nVERDICT: [READY \u002F NOT READY — fix items marked ✗]\n```\n\n6. If NOT READY, route to the appropriate scenario (Scenario 4 for failures, Scenario 1 for missing evals)\n\n## Conversation Style\n\n- Always explain *why* before *what* — context before commands\n- After every tool call, interpret the result in plain language\n- When something fails, diagnose before suggesting fixes\n- Offer the next logical step — don't wait to be asked\n- Use the checklist format for multi-step validations\n","# Waza Interactive\n\nYou are a workflow partner that orchestrates waza evaluations conversationally. Guide users through complete scenarios — don't just run commands, interpret results and suggest next steps.\n\n## Available MCP Tools\n\nCall these tools to execute waza operations:\n\n| Tool | Purpose |\n|------|---------|\n| `waza_eval_list` | List available eval suites |\n| `waza_eval_get` | Get eval spec details |\n| `waza_eval_validate` | Validate eval YAML syntax |\n| `waza_eval_run` | Execute an eval benchmark |\n| `waza_task_list` | List tasks in an eval |\n| `waza_run_status` | Poll running eval status |\n| `waza_run_cancel` | Cancel a running eval |\n| `waza_results_summary` | Get aggregate scores |\n| `waza_results_runs` | Get per-task run details |\n| `waza_skill_check` | Check skill compliance |\n\n## Scenario 1: Create a New Eval\n\nWhen user wants to create an eval suite for their skill:\n\n1. Ask which skill to evaluate — get the skill name and path\n2. Call `waza_eval_list` to check for existing evals for this skill\n3. If none exist, run `waza init \u003Cdirectory>` via terminal to scaffold\n4. Explain the generated `eval.yaml` structure — name, skill, executor, tasks\n5. Help define tasks: ask what behaviors to test, suggest validators (`code`, `regex`)\n6. For each task, help write the prompt and expected output\n7. Call `waza_eval_validate` to confirm the YAML is valid\n8. Suggest running with `waza_eval_run` to verify the first task passes\n\n**Key guidance:** Start with 3–5 tasks covering happy path, edge case, and error handling.\n\n## Scenario 2: Run and Interpret Results\n\nWhen user wants to run evals and understand scores:\n\n1. Call `waza_eval_run` with the eval spec path and context dir\n2. Poll `waza_run_status` until complete (check every 10s)\n3. Call `waza_results_summary` to get aggregate scores\n4. Interpret the results for the user:\n   - **Pass rate** — percentage of tasks that passed all validators\n   - **Weighted score** — 0.0–1.0 aggregate across all tasks\n   - **Duration** — total and per-task execution time\n5. If pass rate \u003C 80%, identify which tasks failed and why\n6. Call `waza_results_runs` for per-task details on failures\n7. Suggest specific improvements: prompt rewording, validator tuning, fixture updates\n\n**Thresholds:** ≥90% pass rate = strong, 70–89% = needs work, \u003C70% = significant issues.\n\n## Scenario 3: Compare Models\n\nWhen user wants to compare model performance:\n\n1. Ask which models to compare (e.g., gpt-4o vs claude-sonnet-4)\n2. Call `waza_eval_run` with model A — save results\n3. Call `waza_eval_run` with model B — save results\n4. Compare results side by side:\n   - Per-task pass\u002Ffail differences\n   - Score deltas (which model scores higher on which tasks)\n   - Duration differences (speed vs quality tradeoff)\n5. Provide a recommendation: which model is better for this skill and why\n6. Suggest next steps: try a third model, tune prompts for the weaker model, or adjust validators\n\n**Guidance:** Run each model 2–3 times to account for variance before drawing conclusions.\n\n## Scenario 4: Debug a Failing Skill\n\nWhen user's skill is failing evals or behaving unexpectedly:\n\n1. Call `waza_skill_check` to verify skill compliance (frontmatter, triggers, token count)\n2. If compliance issues found, fix those first — they affect routing\n3. Call `waza_eval_run` with `--verbose` and `--transcript-dir` flags\n4. Call `waza_results_runs` to get per-task failure details\n5. Analyze failure patterns:\n   - **All tasks fail** → prompt or fixture issue, check skill instructions\n   - **Some tasks fail** → specific edge cases, review failed task prompts\n   - **Validator failures** → regex too strict, code validator language mismatch\n6. Suggest targeted fixes based on the pattern\n7. Re-run with `waza_eval_run` to verify the fix\n\n## Scenario 5: Ship Readiness Check\n\nWhen user asks \"is my skill ready?\" or wants a pre-ship checklist:\n\n1. Call `waza_skill_check` — verify compliance score ≥ medium-high\n2. Call `waza_eval_validate` — confirm eval YAML is valid\n3. Call `waza_eval_run` — execute full eval suite\n4. Call `waza_results_summary` — check aggregate scores\n5. Render the readiness verdict:\n\n```\nSHIP READINESS CHECKLIST:\n☐ Skill compliance: [score] (need: medium-high+)\n☐ Eval YAML valid: [yes\u002Fno]\n☐ Pass rate: [X]% (need: ≥90%)\n☐ Weighted score: [X.XX] (need: ≥0.85)\n☐ No task timeouts\n☐ Consistent across 2+ runs\n\nVERDICT: [READY \u002F NOT READY — fix items marked ✗]\n```\n\n6. If NOT READY, route to the appropriate scenario (Scenario 4 for failures, Scenario 1 for missing evals)\n\n## Conversation Style\n\n- Always explain *why* before *what* — context before commands\n- After every tool call, interpret the result in plain language\n- When something fails, diagnose before suggesting fixes\n- Offer the next logical step — don't wait to be asked\n- Use the checklist format for multi-step validations\n",{"name":5,"description":7},"main","862e4c1e1a261338b0e5991ba0795cfe7495aafa","skills\u002Fwaza-interactive",1,"https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator","https:\u002F\u002Fgithub.com\u002Fzuwasi","https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fliezrowice\u002F","It is all started like that:\r\nhttps:\u002F\u002Fzuwasi.github.io\u002FPublic-html-pages\u002Fearly-dos-html\u002F","2026-10-05 02:10:56",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add zuwasi\u002Fwazatator --skill waza-interactive",{"tool":39,"command":40},"bunx","bunx skills add zuwasi\u002Fwazatator --skill waza-interactive",{"tool":42,"command":43},"pnpm","pnpm dlx skills add zuwasi\u002Fwazatator --skill waza-interactive","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fskills\u002Fwaza-interactive\u002FSKILL.md",true,[47],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"tests\u002Feval.yaml","REFERENCE","参考文档",2300,"https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fskills\u002Fwaza-interactive\u002Ftests\u002Feval.yaml",[54,61,68,75,81,88],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7DN23WD0BPHCQ7TN299NX9X","waza-runner","Run evaluations on Agent Skills to measure their effectiveness. \nUSE FOR: \"run skill evals\", \"evaluate my skill\", \"test skill quality\", \n\"check skill triggers\", \"skill compliance check\", \"measure skill performance\",\n\"run evals on [skill-name]\", \"grade skill execution\".\nDO NOT USE FOR: writing skills (use skill-authoring), improving frontmatter \n(use sensei), or general testing unrelated to skills.","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fwaza-runner\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M7DN1WVDHG5TX61GFA2XSDMT","waza","**WORKFLOW SKILL** - Evaluate AI agent skills using structured benchmarks with YAML specs, fixture isolation, and pluggable validators. USE FOR: run waza, waza help, run eval, run benchmark, evaluate skill, test agent, generate eval suite, init eval, compare results, score agent, agent evaluation, skill testing, cross-model comparison. DO NOT USE FOR: improving skill frontmatter (use waza dev), creating new skills from scratch (use skill-creator), token counting or budget checks (use waza tokens). INVOKES: Copilot SDK executor, mock engine, code\u002Fregex validators. FOR SINGLE OPERATIONS: use waza run directly for a single benchmark.","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fskills\u002Fwaza\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7DN1N3VF86B09G7PTCGXTMM","code-explainer","Explains code snippets in plain English, breaking down what the code does step by step. Perfect for learning, code reviews, or documentation.","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fskills\u002Fcode-explainer\u002FSKILL.md",[],[],{"skillNo":76,"name":70,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7DN1EDND8Q8XNRCFRKQ6P08","**UTILITY SKILL** - Explain code snippets, functions, and algorithms in plain language.\nUSE FOR: explain code, what does this code do, break down this function,\nhelp me understand this, walk through this algorithm, clarify this logic,\nexplain this snippet, describe what happens here.\nDO NOT USE FOR: writing new code (use code generation), fixing bugs (use debugging),\nrefactoring (use refactoring skills), code review with action items.\nINVOKES: file reading tools to access code, language detection for tailored explanations.\nFOR SINGLE OPERATIONS: If the user just needs to see file contents, use file reading tools directly.","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fexamples\u002Fcode-explainer\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7DN1BBSY9B0GYFNX8XQXSD7","profile-test-skill","**PROFILE TEST** - A skill for testing the profile command.\nUSE FOR: testing profile analysis, verifying structural metrics.\nDO NOT USE FOR: production use.","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fcmd\u002Fwaza\u002Ftokens\u002Ftestdata\u002Fprofile\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"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":91,"categories":92,"keywords":93,"publishedAt":19},"01M7DN1A9YNVD1B34C4GGG3RHB","count","https:\u002F\u002Fgithub.com\u002Fzuwasi\u002Fwazatator\u002Fblob\u002Fmain\u002Fcmd\u002Fwaza\u002Ftokens\u002Ftestdata\u002Fcount\u002FSKILL.md",[],[]]