{"id":"40f3af48-a458-472f-9e95-0de1288f124b","shortId":"kkN6rw","kind":"skill","title":"diagnose","tagline":"Perform a systematic diagnostic scan of an AI workflow across 5 quality dimensions — prompt quality, context efficiency, tool health, architecture fitness, and safety — producing a scored report with prioritized remediation actions.","description":"# AI Workflow Diagnostics\n\nYou are a systematic AI workflow auditor. Perform a diagnostic scan across 5 dimensions. For each dimension, score 1–5 and provide specific findings.\n\n## Dimension 1: Prompt Quality (1–5)\n\nEvaluate:\n\n- Structure (role, context, instructions, output zones)\n- Output schema definition (explicit vs. implicit)\n- Instruction clarity (specific vs. vague)\n- Edge case handling (addressed vs. ignored)\n- Anti-patterns (wall of text, contradictions, implicit format)\n\n## Dimension 2: Context Efficiency (1–5)\n\nEvaluate:\n\n- Context budget allocation (planned vs. ad-hoc)\n- Attention gradient awareness (critical info at start/end)\n- Context window utilization (efficient vs. wasteful)\n- State management (explicit vs. implicit)\n- Memory strategy (appropriate for conversation length)\n\n## Dimension 3: Tool Health (1–5)\n\nEvaluate:\n\n- Tool count (3–7 ideal, 13+ problematic)\n- Description quality (specific vs. vague)\n- Error handling (graceful vs. none)\n- Schema completeness (input/output/error defined)\n- Idempotency (safe to retry vs. side-effect prone)\n- **Scope attribution**: Distinguish project-configured tools (custom scripts, project MCP servers) from agent-level tools (built-in IDE tools, global MCP servers). Only flag tool overhead for tools the project can actually control.\n\n## Dimension 4: Architecture Fitness (1–5)\n\nEvaluate:\n\n- Topology appropriateness (single vs. multi-agent justified)\n- Agent boundaries (clear vs. overlapping)\n- Handoff protocols (structured vs. ad-hoc)\n- Observability (decisions logged vs. black box)\n- Cost awareness (budgeted vs. unbounded)\n\n## Dimension 5: Safety & Reliability (1–5)\n\nEvaluate:\n\n- Input validation (present vs. absent)\n- Output filtering (PII, content policy) — scope contextually: data between a user's own frontend and backend is lower risk than data exposed to external services\n- Cost controls (ceilings set vs. unbounded)\n- Error recovery (fallbacks vs. crash)\n- Evaluation strategy (golden tests vs. \"it seems to work\")\n\n## Diagnostic Report Format\n\n```text\n╔══════════════════════════════════════╗\n║          WORKFLOW DIAGNOSTIC        ║\n╠══════════════════════════════════════╣\n║ Prompt Quality      ████░  4/5      ║\n║ Context Efficiency   ███░░  3/5      ║\n║ Tool Health          ██░░░  2/5      ║\n║ Architecture         ████░  4/5      ║\n║ Safety & Reliability ██░░░  2/5      ║\n╠══════════════════════════════════════╣\n║ Overall Score:       15/25           ║\n╚══════════════════════════════════════╝\n\nCRITICAL FINDINGS:\n1. [Most severe issue — immediate action needed]\n2. [Second most severe]\n3. [Third]\n\nRECOMMENDED ACTIONS:\n1. [Specific remediation for finding #1]\n2. [Specific remediation for finding #2]\n3. [Specific remediation for finding #3]\n```\n\n## Scoring Guide\n\n| Score | Meaning                | Recommended Action                        |\n|-------|------------------------|-------------------------------------------|\n| 5     | Production-excellent   | No action needed                          |\n| 4     | Good with minor gaps   | Polish prompt clarity or output schema    |\n| 3     | Functional but risky   | Add error handling or reduce complexity   |\n| 2     | Significant issues     | Immediate attention — add retries/guards  |\n| 1     | Broken or missing      | Rebuild from scratch with clear structure |\n\n## Usage\n\nInvoke this skill when you want to:\n\n- Find hidden problems before a workflow goes to production\n- Audit an existing agent for quality and reliability\n- Get a prioritized remediation plan with concrete next steps\n- Health-check a workflow after significant changes\n\nProvide the workflow description, prompt text, tool list, or agent configuration as context. The more detail you provide, the more precise the findings.","tags":["diagnose","awesome","copilot","github","agent-skills","agents","custom-agents","github-copilot","hacktoberfest","prompt-engineering"],"capabilities":["skill","source-github","skill-diagnose","topic-agent-skills","topic-agents","topic-awesome","topic-custom-agents","topic-github-copilot","topic-hacktoberfest","topic-prompt-engineering"],"categories":["awesome-copilot"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/github/awesome-copilot/diagnose","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add github/awesome-copilot","source_repo":"https://github.com/github/awesome-copilot","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 33270 github stars · SKILL.md body (3,747 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T18:52:10.087Z","embedding":null,"createdAt":"2026-05-06T00:52:18.769Z","updatedAt":"2026-05-18T18:52:10.087Z","lastSeenAt":"2026-05-18T18:52:10.087Z","tsv":"'1':54,61,64,103,142,215,253,331,346,351,405 '13':150 '15/25':328 '2':100,338,352,357,398 '2/5':320,325 '3':139,147,342,358,363,388 '3/5':317 '4':212,377 '4/5':314,322 '5':12,48,55,65,104,143,216,250,254,370 '7':148 'absent':260 'across':11,47 'action':32,336,345,369,375 'actual':209 'ad':112,236 'ad-hoc':111,235 'add':392,403 'address':87 'agent':189,224,226,435,466 'agent-level':188 'ai':9,33,40 'alloc':108 'anti':91 'anti-pattern':90 'appropri':134,219 'architectur':21,213,321 'attent':114,402 'attribut':176 'audit':432 'auditor':42 'awar':116,245 'backend':276 'black':242 'boundari':227 'box':243 'broken':406 'budget':107,246 'built':193 'built-in':192 'case':85 'ceil':288 'chang':456 'check':451 'clariti':80,384 'clear':228,413 'complet':163 'complex':397 'concret':446 'configur':180,467 'content':264 'context':17,69,101,106,121,315,469 'contextu':267 'contradict':96 'control':210,287 'convers':136 'cost':244,286 'count':146 'crash':296 'critic':117,329 'custom':182 'data':268,281 'decis':239 'defin':165 'definit':75 'descript':152,460 'detail':472 'diagnos':1 'diagnost':5,35,45,306,311 'dimens':14,49,52,60,99,138,211,249 'distinguish':177 'edg':84 'effect':173 'effici':18,102,124,316 'error':157,292,393 'evalu':66,105,144,217,255,297 'excel':373 'exist':434 'explicit':76,129 'expos':282 'extern':284 'fallback':294 'filter':262 'find':59,330,350,356,362,423,479 'fit':22,214 'flag':201 'format':98,308 'frontend':274 'function':389 'gap':381 'get':440 'global':197 'goe':429 'golden':299 'good':378 'grace':159 'gradient':115 'guid':365 'handl':86,158,394 'handoff':231 'health':20,141,319,450 'health-check':449 'hidden':424 'hoc':113,237 'ide':195 'ideal':149 'idempot':166 'ignor':89 'immedi':335,401 'implicit':78,97,131 'info':118 'input':256 'input/output/error':164 'instruct':70,79 'invok':416 'issu':334,400 'justifi':225 'length':137 'level':190 'list':464 'log':240 'lower':278 'manag':128 'mcp':185,198 'mean':367 'memori':132 'minor':380 'miss':408 'multi':223 'multi-ag':222 'need':337,376 'next':447 'none':161 'observ':238 'output':71,73,261,386 'overal':326 'overhead':203 'overlap':230 'pattern':92 'perform':2,43 'pii':263 'plan':109,444 'polici':265 'polish':382 'precis':477 'present':258 'priorit':30,442 'problem':425 'problemat':151 'produc':25 'product':372,431 'production-excel':371 'project':179,184,207 'project-configur':178 'prompt':15,62,312,383,461 'prone':174 'protocol':232 'provid':57,457,474 'qualiti':13,16,63,153,313,437 'rebuild':409 'recommend':344,368 'recoveri':293 'reduc':396 'reliabl':252,324,439 'remedi':31,348,354,360,443 'report':28,307 'retri':169 'retries/guards':404 'risk':279 'riski':391 'role':68 'safe':167 'safeti':24,251,323 'scan':6,46 'schema':74,162,387 'scope':175,266 'score':27,53,327,364,366 'scratch':411 'script':183 'second':339 'seem':303 'server':186,199 'servic':285 'set':289 'sever':333,341 'side':172 'side-effect':171 'signific':399,455 'singl':220 'skill':418 'skill-diagnose' 'source-github' 'specif':58,81,154,347,353,359 'start/end':120 'state':127 'step':448 'strategi':133,298 'structur':67,233,414 'systemat':4,39 'test':300 'text':95,309,462 'third':343 'tool':19,140,145,181,191,196,202,205,318,463 'topic-agent-skills' 'topic-agents' 'topic-awesome' 'topic-custom-agents' 'topic-github-copilot' 'topic-hacktoberfest' 'topic-prompt-engineering' 'topolog':218 'unbound':248,291 'usag':415 'user':271 'util':123 'vagu':83,156 'valid':257 'vs':77,82,88,110,125,130,155,160,170,221,229,234,241,247,259,290,295,301 'wall':93 'want':421 'wast':126 'window':122 'work':305 'workflow':10,34,41,310,428,453,459 'zone':72","prices":[{"id":"b790d822-ada2-4ea2-b50c-e38fa823e9a8","listingId":"40f3af48-a458-472f-9e95-0de1288f124b","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"github","category":"awesome-copilot","install_from":"skills.sh"},"createdAt":"2026-05-06T00:52:18.769Z"}],"sources":[{"listingId":"40f3af48-a458-472f-9e95-0de1288f124b","source":"github","sourceId":"github/awesome-copilot/diagnose","sourceUrl":"https://github.com/github/awesome-copilot/tree/main/skills/diagnose","isPrimary":false,"firstSeenAt":"2026-05-06T00:52:18.769Z","lastSeenAt":"2026-05-18T18:52:10.087Z"}],"details":{"listingId":"40f3af48-a458-472f-9e95-0de1288f124b","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"github","slug":"diagnose","github":{"repo":"github/awesome-copilot","stars":33270,"topics":["agent-skills","agents","ai","awesome","custom-agents","github-copilot","hacktoberfest","prompt-engineering"],"license":"mit","html_url":"https://github.com/github/awesome-copilot","pushed_at":"2026-05-18T01:26:59Z","description":"Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.","skill_md_sha":"7f11abaf71b96316b9e3ed18ebe6dc414deeb086","skill_md_path":"skills/diagnose/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/github/awesome-copilot/tree/main/skills/diagnose"},"layout":"multi","source":"github","category":"awesome-copilot","frontmatter":{"name":"diagnose","description":"Perform a systematic diagnostic scan of an AI workflow across 5 quality dimensions — prompt quality, context efficiency, tool health, architecture fitness, and safety — producing a scored report with prioritized remediation actions."},"skills_sh_url":"https://skills.sh/github/awesome-copilot/diagnose"},"updatedAt":"2026-05-18T18:52:10.087Z"}}