{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-cnfjlhj--ai-collab-playbook","slug":"cnfjlhj--ai-collab-playbook","name":"ai-collab-playbook","type":"repo","url":"https://github.com/cnfjlhj/ai-collab-playbook/blob/main/docs/phd-ai-collab.md","page_url":"https://unfragile.ai/cnfjlhj--ai-collab-playbook","categories":["prompt-engineering"],"tags":["ai","ai-collaboration","ai-workflow","claude-code","codex","coding-agent","coding-workflow","playbook","productivity","prompts","research-workflow","writing-workflow"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-cnfjlhj--ai-collab-playbook__cap_0","uri":"capability://text.generation.language.structured.prompt.template.system.for.ai.collaboration","name":"structured-prompt-template-system-for-ai-collaboration","description":"Provides a reusable prompt template framework that decomposes complex research, writing, and coding tasks into structured sections (context, constraints, examples, output format). Templates are designed to be chained together and adapted across different AI models (Claude, GPT, Codex) by maintaining consistent instruction patterns and role definitions that improve consistency and reproducibility across multi-turn conversations.","intents":["I want to create consistent prompts for my research workflow that I can reuse across different AI models without rewriting them each time","I need to structure complex coding tasks so the AI understands my constraints and produces code that matches my project's patterns","I want to build a library of prompt templates that my team can share and iterate on collaboratively"],"best_for":["researchers conducting literature reviews and synthesis with AI assistance","solo developers building coding workflows with multiple AI agents","teams standardizing AI interaction patterns across projects"],"limitations":["Templates are model-agnostic but may require tuning for specific model versions (Claude 3 vs Claude 2 behave differently with identical prompts)","No built-in version control or template inheritance — requires manual updates across copies","Effectiveness depends on user's ability to write clear constraints; poorly specified templates degrade output quality"],"requires":["Access to at least one LLM API (OpenAI, Anthropic, or local Codex)","Markdown editor or IDE for viewing/editing prompt templates","Basic understanding of prompt engineering principles"],"input_types":["markdown text","code snippets","research papers or articles","task descriptions"],"output_types":["structured prompts","code generation instructions","research synthesis frameworks","writing guidelines"],"categories":["text-generation-language","prompt-engineering"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_1","uri":"capability://planning.reasoning.agent.role.definition.framework.for.multi.turn.collaboration","name":"agent-role-definition-framework-for-multi-turn-collaboration","description":"Defines a system for assigning specific roles and responsibilities to AI agents within multi-turn conversations (e.g., 'code reviewer', 'research synthesizer', 'writing editor'). Each role includes explicit behavioral rules, scope boundaries, and interaction patterns that persist across conversation turns, enabling the AI to maintain consistent context and decision-making authority without requiring full context re-specification in each message.","intents":["I want to assign specific responsibilities to Claude so it acts as a code reviewer without drifting into unrelated suggestions","I need the AI to maintain a consistent persona and expertise level across a long research collaboration session","I want to define what the AI should and shouldn't do in my workflow to prevent hallucinations or out-of-scope suggestions"],"best_for":["researchers managing long-form literature synthesis with AI assistance over multiple sessions","developers building coding agents that need to maintain consistent code quality standards","teams implementing AI-assisted code review pipelines with defined approval workflows"],"limitations":["Role definitions are conversation-scoped; no persistent role memory across separate chat sessions without manual re-specification","Overly restrictive role definitions can cause the AI to refuse helpful suggestions outside its defined scope","Role conflicts emerge in multi-agent scenarios where agents have overlapping responsibilities — requires explicit arbitration rules"],"requires":["LLM with strong instruction-following capability (Claude 3+, GPT-4, or equivalent)","Understanding of your task domain to define meaningful role boundaries","Ability to write clear behavioral rules in natural language"],"input_types":["role descriptions (text)","behavioral rules (text)","scope boundaries (text)"],"output_types":["agent instructions","conversation guidelines","role-specific response formats"],"categories":["planning-reasoning","prompt-engineering"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_2","uri":"capability://text.generation.language.research.workflow.prompt.orchestration.for.literature.synthesis","name":"research-workflow-prompt-orchestration-for-literature-synthesis","description":"Provides a sequence of specialized prompts designed to guide AI through research tasks: paper summarization, cross-paper synthesis, gap identification, and argument extraction. Each prompt is optimized for a specific research subtask and includes examples of desired output formats, enabling researchers to decompose literature review work into AI-assisted steps that maintain academic rigor and citation accuracy across multiple sources.","intents":["I want to quickly summarize 20 research papers and extract key arguments without manually reading each one","I need to identify gaps in the literature and synthesize findings across papers to support my thesis","I want the AI to extract citations and maintain academic integrity while helping me organize research"],"best_for":["PhD students and researchers conducting systematic literature reviews","academic teams synthesizing findings across multiple papers for survey articles","researchers building literature maps and identifying research gaps"],"limitations":["AI may misinterpret or oversimplify complex academic arguments — requires human verification of extracted claims","Citation accuracy depends on source document quality; OCR errors or formatting issues propagate through synthesis","Prompt sequence assumes sequential execution; parallel or non-linear research workflows require manual adaptation","Limited to text-based papers; cannot process figures, tables, or supplementary materials without additional preprocessing"],"requires":["Access to full-text research papers (PDF or text format)","LLM with strong reading comprehension (Claude 3+, GPT-4)","Familiarity with your research domain to validate AI-extracted insights","Citation management tool or system for organizing extracted references"],"input_types":["research papers (PDF or text)","paper abstracts","research questions (text)"],"output_types":["paper summaries","synthesis documents","gap analysis reports","structured argument extraction"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_3","uri":"capability://text.generation.language.writing.workflow.prompt.chain.for.iterative.drafting","name":"writing-workflow-prompt-chain-for-iterative-drafting","description":"Provides a structured sequence of prompts for writing tasks: outline generation, draft creation, editing passes (clarity, tone, structure), and final polish. Each step includes specific feedback mechanisms and revision instructions that guide the AI to improve writing iteratively. The workflow maintains document context across steps, allowing writers to refine arguments and style without restarting from scratch.","intents":["I want to outline a complex article and have the AI help me draft sections while maintaining my voice and argument structure","I need to revise my writing for clarity and tone without losing my original ideas","I want to generate multiple writing variations and select the best one for my audience"],"best_for":["technical writers producing documentation and guides","content creators writing articles, blog posts, and long-form content","non-native English speakers improving writing clarity with AI assistance"],"limitations":["AI may homogenize writing style across iterations, reducing unique voice — requires explicit style preservation instructions","Prompt chain assumes linear workflow; non-linear editing (jumping between sections) requires manual context management","Tone and audience assumptions must be explicitly specified; generic prompts produce generic writing","No built-in plagiarism detection or originality checking — requires external tools for verification"],"requires":["LLM with strong writing capability (Claude 3+, GPT-4)","Clear understanding of target audience and writing goals","Ability to provide feedback on AI-generated drafts to guide iterations"],"input_types":["writing prompts (text)","outlines (text or structured)","draft sections (text)","style guidelines (text)"],"output_types":["article outlines","draft sections","revised documents","editing suggestions"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_4","uri":"capability://code.generation.editing.coding.workflow.prompt.system.with.code.quality.rules","name":"coding-workflow-prompt-system-with-code-quality-rules","description":"Defines a set of prompts for code generation, review, and refactoring that embed project-specific coding standards, architecture patterns, and quality constraints. Prompts include examples of desired code style, error handling patterns, and testing requirements, enabling AI code generation to align with team standards. The system supports both single-file generation and multi-file architectural changes by maintaining context about project structure and dependencies.","intents":["I want the AI to generate code that matches my project's style, patterns, and quality standards without manual refactoring","I need to review code changes across multiple files while ensuring they follow our architecture and testing requirements","I want to refactor legacy code while maintaining compatibility and improving readability according to our standards"],"best_for":["development teams standardizing code generation across projects","solo developers building coding agents that understand their codebase patterns","teams implementing AI-assisted code review with defined quality gates"],"limitations":["Code quality rules must be explicitly specified; implicit team conventions are not automatically captured","AI may generate syntactically correct but semantically incorrect code — requires human code review before merging","Multi-file changes require explicit dependency tracking; AI may miss cross-file impacts","Prompt-based approach scales poorly for very large codebases (>100k LOC) without codebase indexing"],"requires":["LLM with strong code understanding (Claude 3.5 Sonnet, GPT-4, or equivalent)","Documented coding standards and architecture patterns for your project","Version control system (Git) for tracking AI-generated changes","Testing framework for validating AI-generated code"],"input_types":["code snippets","project structure descriptions","coding standards (text or examples)","feature requirements (text)"],"output_types":["generated code","code review comments","refactoring suggestions","test cases"],"categories":["code-generation-editing","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_5","uri":"capability://text.generation.language.reusable.skill.library.for.prompt.composition","name":"reusable-skill-library-for-prompt-composition","description":"Provides a collection of modular, reusable prompt components (skills) that can be combined to build complex AI workflows. Skills are organized by function (e.g., 'extract key points', 'generate examples', 'identify contradictions') and include clear input/output specifications, enabling users to compose custom workflows by chaining skills together without writing prompts from scratch.","intents":["I want to build a custom research workflow by combining existing prompt skills rather than writing new prompts","I need to reuse successful prompt patterns across different projects and tasks","I want to share prompt components with my team so we don't duplicate effort"],"best_for":["teams building multiple AI workflows and wanting to avoid prompt duplication","researchers creating custom analysis pipelines from modular components","organizations standardizing AI collaboration patterns across departments"],"limitations":["Skill composition requires understanding of input/output contracts — poorly matched skills produce degraded results","No automatic skill discovery or recommendation — users must manually select and combine skills","Skill library requires maintenance as new patterns emerge or old patterns become obsolete","Context loss between skill boundaries may require explicit context-passing mechanisms"],"requires":["Documentation of available skills and their input/output specifications","LLM capable of executing chained prompts with context preservation","Version control or skill registry for managing skill library"],"input_types":["skill definitions (text)","skill parameters (structured data)"],"output_types":["composed workflows","skill chains","custom prompts"],"categories":["text-generation-language","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_6","uri":"capability://text.generation.language.multi.model.prompt.adaptation.for.cross.platform.ai.collaboration","name":"multi-model-prompt-adaptation-for-cross-platform-ai-collaboration","description":"Provides guidance for adapting prompts across different LLM platforms (Claude, GPT, Codex, local models) by documenting model-specific behaviors, instruction formats, and output patterns. The playbook includes examples of how to adjust prompts for different model capabilities (e.g., Claude's strong reasoning vs GPT's broader knowledge) while maintaining consistent intent, enabling users to switch models or use multiple models in parallel without complete prompt rewrites.","intents":["I want to use Claude for reasoning-heavy tasks and GPT for knowledge-heavy tasks without maintaining separate prompt libraries","I need to migrate my prompts from OpenAI to Anthropic without rewriting everything","I want to run the same workflow across multiple models to compare outputs and select the best result"],"best_for":["teams using multiple LLM providers and wanting to standardize prompts","developers building model-agnostic AI workflows","researchers comparing model capabilities on specific tasks"],"limitations":["Model-specific adaptations require knowledge of each model's strengths and quirks — generic adaptations may not optimize for specific models","Prompt compatibility is not guaranteed across model versions; updates to Claude or GPT may require re-tuning","Cost and latency vary significantly across models; prompt optimization for one model may not be cost-optimal for another","Some model-specific features (e.g., vision, function calling) cannot be easily abstracted across platforms"],"requires":["API access to multiple LLM providers (OpenAI, Anthropic, etc.)","Understanding of model-specific instruction formats and capabilities","Testing framework for validating prompt behavior across models"],"input_types":["base prompts (text)","model specifications (text)"],"output_types":["model-adapted prompts","compatibility guides","model comparison reports"],"categories":["text-generation-language","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cnfjlhj--ai-collab-playbook__cap_7","uri":"capability://memory.knowledge.collaborative.ai.session.management.with.context.preservation","name":"collaborative-ai-session-management-with-context-preservation","description":"Provides patterns for managing long-form AI collaboration sessions that maintain context, conversation history, and task state across multiple turns without losing information or requiring full context re-specification. Includes techniques for summarizing conversation history, managing token limits, and preserving key decisions and constraints across session boundaries, enabling researchers and developers to maintain productive AI partnerships over extended periods.","intents":["I want to continue a research collaboration with Claude across multiple sessions without losing context or having to re-explain my project","I need to manage long conversations that exceed token limits while preserving important decisions and constraints","I want to extract and reuse insights from past AI collaborations in new projects"],"best_for":["researchers conducting multi-week literature reviews with AI assistance","developers building complex features with AI pair programming over extended periods","teams maintaining long-term AI-assisted workflows that span multiple sessions"],"limitations":["Context summarization may lose nuance or important details — requires periodic human review of summaries","Token limit management requires active monitoring and manual intervention; no automatic optimization","Session state persistence requires external storage (database, files); no built-in session management","Conversation history grows unbounded; archival and cleanup require manual processes"],"requires":["LLM with strong summarization capability (Claude 3+, GPT-4)","External storage for session state and conversation history","Discipline in documenting key decisions and constraints for future reference"],"input_types":["conversation history (text)","session summaries (text)","task state (structured data)"],"output_types":["session summaries","context preservation documents","decision logs"],"categories":["memory-knowledge","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":37,"verified":false,"data_access_risk":"high","permissions":["Access to at least one LLM API (OpenAI, Anthropic, or local Codex)","Markdown editor or IDE for viewing/editing prompt templates","Basic understanding of prompt engineering principles","LLM with strong instruction-following capability (Claude 3+, GPT-4, or equivalent)","Understanding of your task domain to define meaningful role boundaries","Ability to write clear behavioral rules in natural language","Access to full-text research papers (PDF or text format)","LLM with strong reading comprehension (Claude 3+, GPT-4)","Familiarity with your research domain to validate AI-extracted insights","Citation management tool or system for organizing extracted references"],"failure_modes":["Templates are model-agnostic but may require tuning for specific model versions (Claude 3 vs Claude 2 behave differently with identical prompts)","No built-in version control or template inheritance — requires manual updates across copies","Effectiveness depends on user's ability to write clear constraints; poorly specified templates degrade output quality","Role definitions are conversation-scoped; no persistent role memory across separate chat sessions without manual re-specification","Overly restrictive role definitions can cause the AI to refuse helpful suggestions outside its defined scope","Role conflicts emerge in multi-agent scenarios where agents have overlapping responsibilities — requires explicit arbitration rules","AI may misinterpret or oversimplify complex academic arguments — requires human verification of extracted claims","Citation accuracy depends on source document quality; OCR errors or formatting issues propagate through synthesis","Prompt sequence assumes sequential execution; parallel or non-linear research workflows require manual adaptation","Limited to text-based papers; cannot process figures, tables, or supplementary materials without additional preprocessing","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.268777742724144,"quality":0.41,"ecosystem":0.6000000000000001,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.3,"quality":0.2,"ecosystem":0.15,"match_graph":0.3,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.549Z","last_scraped_at":"2026-05-03T13:59:55.150Z","last_commit":"2026-04-26T09:41:40Z"},"community":{"stars":255,"forks":17,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=cnfjlhj--ai-collab-playbook","compare_url":"https://unfragile.ai/compare?artifact=cnfjlhj--ai-collab-playbook"}},"signature":"+EAsnpdE7lUP+yav4z2ZKLSXj8gzR4MlcgBNbzbr9gU+KktxmsYOFELkR1A0hV45x/IV+Mamdljus5M7+O/aAQ==","signedAt":"2026-06-20T06:05:18.402Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/cnfjlhj--ai-collab-playbook","artifact":"https://unfragile.ai/cnfjlhj--ai-collab-playbook","verify":"https://unfragile.ai/api/v1/verify?slug=cnfjlhj--ai-collab-playbook","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}