{"id":"680cd1c3-0888-429d-8ed5-ddba96dfa366","shortId":"buR2WY","kind":"skill","title":"oracle-thinkdeep","tagline":"Extended reasoning analysis using sequential thinking. Use for deep exploration, comprehensive analysis, complex decisions, or when you need fresh perspectives on difficult problems.","description":"# ThinkDeep: Extended Reasoning Analysis\n\n## Step 1: Parse Request Parameters\n\n<strategist>\n@agent-oracle\n\nAnalyze the thinkdeep request: $ARGUMENTS\n\n**Parameter Extraction:**\n- **Main prompt**: [extract core thinking request]\n- **Focus areas**: [extract \"architecture, performance, security\" etc]\n- **File references**: [extract \"reference to file.py\" patterns]\n- **Problem context**: [extract additional context provided]\n\n**Parsed Configuration:**\n- Prompt: [core request]\n- Focus areas: [identified areas]\n- File references: [any files mentioned]\n- Problem context: [additional context]\n</strategist>\n\n## Step 2: Deep Sequential Analysis\n\n<strategist>\n@agent-oracle\n\nUse sequential thinking (mcp__sequential-thinking__sequentialthinking) for extended reasoning:\n\n### Thought 1: Context Understanding\nAnalyze the current session context to understand:\n- What has been discussed so far\n- Previous decisions and analyses made\n- User's specific constraints and goals\n- Existing solutions or approaches tried\n\n### Thought 2: Problem Decomposition\nBreak down the core issue into:\n- Primary problem statement\n- Key sub-problems or components\n- Underlying assumptions and constraints\n- Success criteria and desired outcomes\n\n### Thought 3: Assumption Challenge\nQuestion the underlying assumptions:\n- What assumptions are we making?\n- Are these assumptions valid in this context?\n- What if the opposite were true?\n- Alternative perspectives to consider\n\n### Thought 4: Alternative Approaches Exploration\nExplore different solution paths:\n- Conventional approaches and their pros/cons\n- Unconventional or innovative solutions\n- Trade-offs between different options\n- Hybrid or combined approaches\n\n### Thought 5: Context Integration\nIncorporate relevant context:\n- File references and code patterns\n- Previous session insights\n- User's specific technical stack\n- Business or project constraints\n\n### Thought 6: Edge Case Analysis\nIdentify potential issues:\n- What could go wrong with each approach?\n- Edge cases and failure modes\n- Scalability and performance concerns\n- Security and maintenance implications\n\n### Thought 7: Risk Assessment\nEvaluate risks for each option:\n- Implementation complexity\n- Technical debt potential\n- Team skill requirements\n- Timeline and resource impacts\n\n### Thought 8: Solution Synthesis\nCombine insights into recommendations:\n- Primary recommended approach\n- Backup or alternative options\n- Implementation considerations\n- Next steps and validation points\n\nContinue sequential thinking until satisfaction, using:\n- `isRevision: true` to refine previous thoughts\n- `branchFromThought` to explore alternative reasoning paths\n- `needsMoreThoughts: true` to extend analysis if needed\n</strategist>\n\n## Step 3: Critical Evaluation & Synthesis\n\n**Self-Critique Questions:**\n- Does the analysis address the user's specific context?\n- Are the recommendations practical and implementable?\n- Have we considered the most important constraints?\n- Are there any blind spots or missing perspectives?\n\n**Contextual Validation:**\n- Fit with user's technical stack and preferences\n- Alignment with previous decisions in session\n- Practicality given team skills and resources\n- Integration with existing systems or patterns\n\n**Final Synthesis:**\n- **Primary Recommendation**: [best approach with rationale]\n- **Alternative Options**: [backup plans with trade-offs]\n- **Implementation Guidance**: [key considerations and next steps]\n- **Risk Mitigation**: [how to address identified concerns]\n\n---\n\n## Usage Examples\n\n**Basic deep analysis:**\n```\n/atelier-thinkdeep \"Analyze my authentication architecture\"\n```\n\n**With file references:**\n```\n/atelier-thinkdeep \"Evaluate my database schema with reference to models/user.py\"\n```\n\n**With focus areas:**\n```\n/atelier-thinkdeep \"Assess my microservices design focusing on performance and security\"\n```\n\n**Complex architectural decisions:**\n```\n/atelier-thinkdeep \"Should I use GraphQL or REST for my API considering team skills and scalability needs\"\n```\n\n## When to Use ThinkDeep\n\n- **Architecture decisions**: When choosing between major design approaches\n- **Complex problem solving**: Multi-faceted issues requiring deep analysis\n- **Assumption validation**: Challenging conventional wisdom or previous decisions\n- **Alternative exploration**: When stuck and need fresh perspectives\n- **Risk assessment**: Before major implementation decisions\n\n## ThinkDeep vs Challenge\n\n**Use /atelier-thinkdeep**: Deep exploration, comprehensive analysis, alternative discovery, complex decisions\n**Use /atelier-challenge**: Question assumptions, test validity, assess risks, prevent automatic agreement\n\n**Key distinction**: ThinkDeep = deep exploration, Challenge = critical evaluation\n\n## Expected Outcomes\n\nAfter using this command:\n- Comprehensive understanding of the problem space\n- Multiple approaches evaluated with trade-offs\n- Clear recommendation with solid reasoning\n- Risk assessment and mitigation strategies\n- Actionable next steps for implementation","tags":["oracle","thinkdeep","atelier","martinffx","agent-skills","agentic-coding","anthropic","claude-code","claude-skills","code-review","codex","codex-skill"],"capabilities":["skill","source-martinffx","skill-oracle-thinkdeep","topic-agent-skills","topic-agentic-coding","topic-anthropic","topic-claude-code","topic-claude-skills","topic-code-review","topic-codex","topic-codex-skill","topic-opencode","topic-prompt-engineering","topic-sdd","topic-spec-driven-development"],"categories":["atelier"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/martinffx/atelier/oracle-thinkdeep","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add martinffx/atelier","source_repo":"https://github.com/martinffx/atelier","install_from":"skills.sh"}},"qualityScore":"0.461","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 23 github stars · SKILL.md body (5,011 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":"2026-05-18T19:05:23.043Z","embedding":null,"createdAt":"2026-05-10T07:03:11.886Z","updatedAt":"2026-05-18T19:05:23.043Z","lastSeenAt":"2026-05-18T19:05:23.043Z","tsv":"'/atelier-challenge':557 '/atelier-thinkdeep':450,458,470,483,547 '1':32,110 '2':91,143 '3':171,349 '4':201 '5':229 '6':253 '7':281 '8':302 'action':604 'addit':69,88 'address':360,442 'agent':37,96 'agent-oracl':36,95 'agreement':566 'align':397 'altern':196,202,314,338,423,529,552 'analys':129 'analysi':6,15,30,94,256,345,359,449,520,551 'analyz':39,113,451 'api':492 'approach':140,203,210,227,266,311,420,510,588 'architectur':55,454,481,503 'area':53,78,80,469 'argument':43 'assess':283,471,538,562,600 'assumpt':162,172,177,179,185,521,559 'authent':453 'automat':565 'backup':312,425 'basic':447 'best':419 'blind':382 'branchfromthought':335 'break':146 'busi':248 'case':255,268 'challeng':173,523,545,572 'choos':506 'clear':594 'code':238 'combin':226,305 'command':580 'complex':16,290,480,511,554 'compon':160 'comprehens':14,550,581 'concern':275,444 'configur':73 'consid':199,374,493 'consider':317,434 'constraint':134,164,251,378 'context':67,70,87,89,111,117,189,230,234,365 'contextu':387 'continu':323 'convent':209,524 'core':49,75,149 'could':261 'criteria':166 'critic':350,573 'critiqu':355 'current':115 'databas':461 'debt':292 'decis':17,127,400,482,504,528,542,555 'decomposit':145 'deep':12,92,448,519,548,570 'design':474,509 'desir':168 'differ':206,222 'difficult':25 'discoveri':553 'discuss':123 'distinct':568 'edg':254,267 'etc':58 'evalu':284,351,459,574,589 'exampl':446 'exist':137,411 'expect':575 'explor':13,204,205,337,530,549,571 'extend':4,28,107,344 'extract':45,48,54,61,68 'facet':516 'failur':270 'far':125 'file':59,81,84,235,456 'file.py':64 'final':415 'fit':389 'focus':52,77,468,475 'fresh':22,535 'given':404 'go':262 'goal':136 'graphql':487 'guidanc':432 'hybrid':224 'identifi':79,257,443 'impact':300 'implement':289,316,371,431,541,608 'implic':279 'import':377 'incorpor':232 'innov':216 'insight':242,306 'integr':231,409 'isrevis':329 'issu':150,259,517 'key':155,433,567 'made':130 'main':46 'mainten':278 'major':508,540 'make':182 'mcp':101 'mention':85 'microservic':473 'miss':385 'mitig':439,602 'mode':271 'models/user.py':466 'multi':515 'multi-facet':514 'multipl':587 'need':21,347,498,534 'needsmorethought':341 'next':318,436,605 'off':220,430,593 'opposit':193 'option':223,288,315,424 'oracl':2,38,97 'oracle-thinkdeep':1 'outcom':169,576 'paramet':35,44 'pars':33,72 'path':208,340 'pattern':65,239,414 'perform':56,274,477 'perspect':23,197,386,536 'plan':426 'point':322 'potenti':258,293 'practic':369,403 'prefer':396 'prevent':564 'previous':126,240,333,399,527 'primari':152,309,417 'problem':26,66,86,144,153,158,512,585 'project':250 'prompt':47,74 'pros/cons':213 'provid':71 'question':174,356,558 'rational':422 'reason':5,29,108,339,598 'recommend':308,310,368,418,595 'refer':60,62,82,236,457,464 'refin':332 'relev':233 'request':34,42,51,76 'requir':296,518 'resourc':299,408 'rest':489 'risk':282,285,438,537,563,599 'satisfact':327 'scalabl':272,497 'schema':462 'secur':57,276,479 'self':354 'self-critiqu':353 'sequenti':8,93,99,103,324 'sequential-think':102 'sequentialthink':105 'session':116,241,402 'skill':295,406,495 'skill-oracle-thinkdeep' 'solid':597 'solut':138,207,217,303 'solv':513 'source-martinffx' 'space':586 'specif':133,245,364 'spot':383 'stack':247,394 'statement':154 'step':31,90,319,348,437,606 'strategi':603 'stuck':532 'sub':157 'sub-problem':156 'success':165 'synthesi':304,352,416 'system':412 'team':294,405,494 'technic':246,291,393 'test':560 'think':9,50,100,104,325 'thinkdeep':3,27,41,502,543,569 'thought':109,142,170,200,228,252,280,301,334 'timelin':297 'topic-agent-skills' 'topic-agentic-coding' 'topic-anthropic' 'topic-claude-code' 'topic-claude-skills' 'topic-code-review' 'topic-codex' 'topic-codex-skill' 'topic-opencode' 'topic-prompt-engineering' 'topic-sdd' 'topic-spec-driven-development' 'trade':219,429,592 'trade-off':218,428,591 'tri':141 'true':195,330,342 'unconvent':214 'under':161,176 'understand':112,119,582 'usag':445 'use':7,10,98,328,486,501,546,556,578 'user':131,243,362,391 'valid':186,321,388,522,561 'vs':544 'wisdom':525 'wrong':263","prices":[{"id":"f241044b-58ca-45ac-b81b-cc17c770309a","listingId":"680cd1c3-0888-429d-8ed5-ddba96dfa366","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"martinffx","category":"atelier","install_from":"skills.sh"},"createdAt":"2026-05-10T07:03:11.886Z"}],"sources":[{"listingId":"680cd1c3-0888-429d-8ed5-ddba96dfa366","source":"github","sourceId":"martinffx/atelier/oracle-thinkdeep","sourceUrl":"https://github.com/martinffx/atelier/tree/main/skills/oracle-thinkdeep","isPrimary":false,"firstSeenAt":"2026-05-10T07:03:11.886Z","lastSeenAt":"2026-05-18T19:05:23.043Z"}],"details":{"listingId":"680cd1c3-0888-429d-8ed5-ddba96dfa366","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"martinffx","slug":"oracle-thinkdeep","github":{"repo":"martinffx/atelier","stars":23,"topics":["agent-skills","agentic-coding","anthropic","claude-code","claude-skills","code-review","codex","codex-skill","opencode","prompt-engineering","sdd","spec-driven-development"],"license":"mit","html_url":"https://github.com/martinffx/atelier","pushed_at":"2026-05-18T06:56:45Z","description":"An atelier for Opencode, Claude Code, and other coding agents: spec-driven workflows, deep thinking, and code quality.","skill_md_sha":"5eaa43e297f2799245806520c5fc06aec28b64eb","skill_md_path":"skills/oracle-thinkdeep/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/martinffx/atelier/tree/main/skills/oracle-thinkdeep"},"layout":"multi","source":"github","category":"atelier","frontmatter":{"name":"oracle-thinkdeep","description":"Extended reasoning analysis using sequential thinking. Use for deep exploration, comprehensive analysis, complex decisions, or when you need fresh perspectives on difficult problems."},"skills_sh_url":"https://skills.sh/martinffx/atelier/oracle-thinkdeep"},"updatedAt":"2026-05-18T19:05:23.043Z"}}