{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-jparkerweb--ai-assistant-prompts","slug":"jparkerweb--ai-assistant-prompts","name":"ai-assistant-prompts","type":"prompt","url":"https://github.com/jparkerweb/ai-assistant-prompts","page_url":"https://unfragile.ai/jparkerweb--ai-assistant-prompts","categories":["prompt-engineering"],"tags":["ai","ai-prompts","ai-workflow","equill-plugin"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-jparkerweb--ai-assistant-prompts__cap_0","uri":"capability://text.generation.language.system.prompt.templating.for.agent.roles","name":"system-prompt-templating-for-agent-roles","description":"Provides pre-written, role-specific system prompts that define agent behavior, constraints, and communication style for different use cases (coding assistant, creative writer, analyst, etc.). Works by offering curated prompt templates that can be directly injected into LLM system contexts or modified for specific agent personalities. Templates encode behavioral guardrails, tone preferences, and domain-specific instructions without requiring prompt engineering from scratch.","intents":["I need to quickly spin up an AI agent with consistent behavior without writing system prompts from scratch","I want to ensure my agent follows specific rules and constraints across multiple conversations","I need different agent personalities for different tasks but want a standardized approach"],"best_for":["developers building multi-agent systems with distinct roles","teams standardizing AI behavior across products","rapid prototypers testing agent configurations"],"limitations":["Templates are generic and may require customization for domain-specific nuances","No built-in versioning or A/B testing framework for prompt variants","Limited to static templates — no dynamic prompt generation based on context"],"requires":["LLM API access (OpenAI, Anthropic, or compatible)","Understanding of how to inject system prompts into your LLM client"],"input_types":["text (agent role/use case selection)"],"output_types":["text (system prompt template)"],"categories":["text-generation-language","prompt-engineering"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_1","uri":"capability://planning.reasoning.agent.behavior.rule.definition","name":"agent-behavior-rule-definition","description":"Encodes explicit behavioral rules and constraints within prompts that govern how agents respond to edge cases, handle errors, manage context limits, and enforce safety boundaries. Rules are expressed as natural language instructions embedded in system prompts, allowing agents to follow deterministic logic without code changes. Patterns include conditional rules (if-then logic), constraint hierarchies, and fallback behaviors.","intents":["I need my agent to refuse certain requests or topics consistently","I want to define what the agent should do when it encounters ambiguous input","I need to enforce rate limits, context window awareness, or resource constraints at the prompt level"],"best_for":["teams building safety-critical agents","developers implementing content moderation at the prompt level","builders needing deterministic agent behavior without custom code"],"limitations":["Rule enforcement depends on LLM compliance — no guarantee rules will be followed","Complex conditional logic becomes hard to maintain in natural language","No runtime monitoring or audit trail for rule violations","Rules are not executable code — they're suggestions to the model"],"requires":["LLM with sufficient instruction-following capability","Understanding of prompt injection risks when rules are user-facing"],"input_types":["text (rule definitions in natural language)"],"output_types":["text (agent responses constrained by rules)"],"categories":["planning-reasoning","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_10","uri":"capability://memory.knowledge.knowledge.grounding.and.source.attribution.prompts","name":"knowledge-grounding-and-source-attribution-prompts","description":"Provides prompt templates that instruct agents to ground responses in provided knowledge bases, cite sources, and distinguish between known facts and speculation. Templates include instructions for referencing specific documents, acknowledging uncertainty, and avoiding hallucination. Implemented as system prompt components that make agents source-aware and fact-conscious.","intents":["I need my agent to cite sources for claims and avoid making up information","I want the agent to distinguish between facts from my knowledge base and general knowledge","I need the agent to acknowledge when it doesn't know something rather than guessing"],"best_for":["teams building knowledge-grounded agents (RAG systems)","organizations with strict factuality requirements","developers creating agents for customer support or documentation"],"limitations":["Source attribution depends on agent compliance — not guaranteed","Agents may still hallucinate or confabulate sources","Distinguishing between knowledge base facts and general knowledge is difficult","No automated verification that cited sources actually support claims"],"requires":["Knowledge base or document collection to ground responses","LLM with good instruction-following for source attribution","Mechanism to provide knowledge base context to the agent (RAG or similar)"],"input_types":["text (knowledge base documents, queries)"],"output_types":["text (source-attributed responses)"],"categories":["memory-knowledge","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_2","uri":"capability://planning.reasoning.multi.agent.interaction.protocol.templates","name":"multi-agent-interaction-protocol-templates","description":"Provides prompt templates that define how multiple agents should communicate, coordinate, and hand off tasks to each other. Templates specify message formats, turn-taking rules, context passing mechanisms, and conflict resolution strategies. Enables orchestration of agent conversations without building custom communication protocols by encoding interaction patterns directly in system prompts.","intents":["I need multiple agents to work together on a complex task with clear handoff points","I want to define how agents should request help from or delegate to other agents","I need agents to maintain shared context and avoid duplicating work"],"best_for":["teams building multi-agent reasoning systems","developers implementing agent orchestration without a dedicated framework","builders prototyping agent collaboration patterns"],"limitations":["No built-in state management — agents must pass context through prompts","Scaling to 3+ agents becomes complex; coordination overhead grows exponentially","No formal verification that agents follow the protocol","Requires careful prompt design to avoid infinite loops or deadlocks"],"requires":["Multiple LLM API endpoints or a single endpoint supporting parallel requests","Mechanism to route messages between agents (custom orchestration layer)"],"input_types":["text (task description, agent roles)"],"output_types":["text (agent responses, coordination messages)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_3","uri":"capability://text.generation.language.domain.specific.agent.persona.library","name":"domain-specific-agent-persona-library","description":"Provides pre-configured agent personas tailored to specific domains (coding, creative writing, data analysis, customer support, etc.) with domain-appropriate vocabulary, reasoning patterns, and response styles. Each persona template includes domain-specific instructions, common task patterns, and expected output formats. Personas are implemented as system prompt variants that can be selected and customized based on the task domain.","intents":["I need a coding assistant that understands architecture patterns and best practices","I want a creative writing agent that maintains consistent voice and style","I need a data analyst agent that explains findings in business terms"],"best_for":["product teams building specialized AI assistants for specific domains","developers creating domain-specific chatbots","non-technical users who want domain expertise without prompt engineering"],"limitations":["Personas are static templates — no learning or adaptation to user preferences over time","Domain coverage is limited to what's in the repository","Persona conflicts may arise when a task spans multiple domains","No mechanism to blend or compose multiple personas"],"requires":["LLM API access","Knowledge of which domain/persona matches your use case"],"input_types":["text (domain selection, task description)"],"output_types":["text (domain-specific response)"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_4","uri":"capability://planning.reasoning.prompt.composition.and.chaining.patterns","name":"prompt-composition-and-chaining-patterns","description":"Provides templates and patterns for composing multiple prompts into chains or workflows where output from one prompt feeds into the next. Patterns include sequential chaining (output → next input), branching (conditional routing), and aggregation (combining multiple outputs). Enables complex reasoning by breaking tasks into prompt-based steps without requiring code-based orchestration.","intents":["I need to break a complex task into steps where each step uses a different prompt","I want to route tasks to different agents based on intermediate results","I need to aggregate outputs from multiple prompts into a final answer"],"best_for":["developers building prompt-based workflows","teams implementing chain-of-thought reasoning at scale","builders prototyping complex reasoning without custom code"],"limitations":["No built-in error handling or retry logic — failures cascade through the chain","Latency compounds with each step (N prompts = N API calls)","Debugging multi-step chains is difficult without logging infrastructure","Context window limits become problematic as chains grow longer"],"requires":["LLM API access with support for sequential requests","Orchestration layer to manage prompt chaining (custom code or framework)"],"input_types":["text (initial task, intermediate results)"],"output_types":["text (final result after all chain steps)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_5","uri":"capability://safety.moderation.safety.and.alignment.constraint.templates","name":"safety-and-alignment-constraint-templates","description":"Provides pre-written constraint prompts that enforce safety boundaries, prevent harmful outputs, and align agent behavior with organizational values. Constraints are expressed as explicit instructions covering topics like bias prevention, factuality requirements, content filtering, and ethical guidelines. Implemented as system prompt components that can be combined with task-specific prompts to create safety-aware agents.","intents":["I need to ensure my agent doesn't generate harmful, biased, or illegal content","I want to enforce factuality and source attribution in agent outputs","I need to align my agent with my organization's ethical guidelines"],"best_for":["teams building customer-facing AI products","organizations with strict compliance or safety requirements","developers implementing content moderation at the prompt level"],"limitations":["Constraint compliance depends on LLM instruction-following — not guaranteed","Adversarial prompts may override safety constraints","No runtime monitoring or audit trail for constraint violations","Constraints may conflict with task requirements (e.g., refusing to discuss sensitive topics)"],"requires":["LLM with strong instruction-following and safety awareness","Understanding of prompt injection and jailbreak risks"],"input_types":["text (constraint definitions, task descriptions)"],"output_types":["text (constrained agent responses)"],"categories":["safety-moderation","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_6","uri":"capability://planning.reasoning.error.handling.and.fallback.prompt.patterns","name":"error-handling-and-fallback-prompt-patterns","description":"Provides prompt templates that define how agents should handle errors, edge cases, and ambiguous inputs. Patterns include graceful degradation (providing partial results when full results aren't possible), fallback behaviors (default actions when primary logic fails), and error recovery (asking for clarification or retrying with different approaches). Implemented as conditional instructions embedded in system prompts.","intents":["I need my agent to handle invalid input gracefully without crashing","I want the agent to ask for clarification when it's uncertain","I need the agent to provide partial results when it can't fully complete a task"],"best_for":["developers building robust, production-ready agents","teams implementing user-facing AI with high reliability requirements","builders creating agents that must handle diverse, unpredictable inputs"],"limitations":["Error handling is prompt-based — no guarantee the agent will follow fallback logic","Complex error scenarios may exceed the agent's reasoning capability","No structured error codes or machine-readable error information","Fallback behaviors may not be appropriate for all error types"],"requires":["LLM with good instruction-following for conditional logic","Monitoring to detect when fallback behaviors are triggered"],"input_types":["text (error scenarios, fallback instructions)"],"output_types":["text (error-aware agent responses)"],"categories":["planning-reasoning","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_7","uri":"capability://planning.reasoning.context.window.management.instructions","name":"context-window-management-instructions","description":"Provides prompt templates that instruct agents how to manage limited context windows, including strategies for summarization, prioritization of important information, and graceful handling of context overflow. Templates encode instructions for agents to recognize when context is running low, decide what to keep vs. discard, and request additional context if needed. Implemented as system prompt guidelines that make agents context-aware.","intents":["I need my agent to work within strict context window limits without losing important information","I want the agent to summarize long conversations to fit within token limits","I need the agent to prioritize recent or important messages when context is full"],"best_for":["developers building long-running agent conversations","teams using smaller or cheaper models with limited context windows","builders implementing agents that must handle large documents or histories"],"limitations":["Context management is heuristic-based — agents may discard important information","No guarantee agents will follow context management instructions","Summarization introduces information loss and potential inaccuracy","Context overflow handling is reactive, not proactive"],"requires":["LLM with understanding of token limits and summarization","Monitoring to track actual token usage"],"input_types":["text (context management instructions, conversation history)"],"output_types":["text (context-aware agent responses)"],"categories":["planning-reasoning","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_8","uri":"capability://planning.reasoning.task.decomposition.and.subtask.prompting","name":"task-decomposition-and-subtask-prompting","description":"Provides prompt templates that teach agents to break complex tasks into subtasks, work through them systematically, and combine results. Templates include instructions for identifying subtasks, prioritizing them, tracking completion, and handling dependencies. Enables agents to tackle complex problems by reasoning about task structure without requiring external task management systems.","intents":["I need my agent to break down complex projects into manageable steps","I want the agent to identify dependencies between subtasks and order them correctly","I need the agent to track progress and handle subtask failures gracefully"],"best_for":["developers building agents for complex problem-solving","teams implementing agents for project planning or code generation","builders creating agents that must handle multi-step reasoning"],"limitations":["Task decomposition quality depends on agent reasoning capability","No guarantee agents will identify all subtasks or dependencies","Tracking subtask completion is prompt-based — no persistent state","Handling subtask failures requires agent reasoning, not guaranteed"],"requires":["LLM with strong reasoning and planning capability","Mechanism to track subtask completion (custom code or external system)"],"input_types":["text (complex task description)"],"output_types":["text (subtask list, execution plan, results)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jparkerweb--ai-assistant-prompts__cap_9","uri":"capability://text.generation.language.output.formatting.and.structure.templates","name":"output-formatting-and-structure-templates","description":"Provides prompt templates that specify exact output formats (JSON, markdown, structured text, etc.) and enforce consistent structure across agent responses. Templates include instructions for organizing information hierarchically, using specific headers or sections, and following formatting conventions. Enables downstream systems to parse agent outputs reliably without post-processing.","intents":["I need my agent to output structured data (JSON, CSV) that my application can parse","I want consistent formatting across all agent responses for better UX","I need the agent to organize complex information in a specific structure"],"best_for":["developers building agent APIs or integrations","teams needing machine-readable agent outputs","builders creating agents that feed into downstream systems"],"limitations":["Format compliance depends on LLM instruction-following — not guaranteed","Complex formats may confuse the agent or reduce output quality","Parsing errors may occur if agent doesn't follow format exactly","Format constraints may limit agent expressiveness"],"requires":["LLM with good instruction-following for structured output","Parser or validation logic to handle format errors"],"input_types":["text (format specifications, task descriptions)"],"output_types":["text (JSON, markdown, CSV, or other structured formats)"],"categories":["text-generation-language","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":29,"verified":false,"data_access_risk":"low","permissions":["LLM API access (OpenAI, Anthropic, or compatible)","Understanding of how to inject system prompts into your LLM client","LLM with sufficient instruction-following capability","Understanding of prompt injection risks when rules are user-facing","Knowledge base or document collection to ground responses","LLM with good instruction-following for source attribution","Mechanism to provide knowledge base context to the agent (RAG or similar)","Multiple LLM API endpoints or a single endpoint supporting parallel requests","Mechanism to route messages between agents (custom orchestration layer)","LLM API access"],"failure_modes":["Templates are generic and may require customization for domain-specific nuances","No built-in versioning or A/B testing framework for prompt variants","Limited to static templates — no dynamic prompt generation based on context","Rule enforcement depends on LLM compliance — no guarantee rules will be followed","Complex conditional logic becomes hard to maintain in natural language","No runtime monitoring or audit trail for rule violations","Rules are not executable code — they're suggestions to the model","Source attribution depends on agent compliance — not guaranteed","Agents may still hallucinate or confabulate sources","Distinguishing between knowledge base facts and general knowledge is difficult","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.10561249512262474,"quality":0.32,"ecosystem":0.52,"match_graph":0.25,"freshness":0.6,"weights":{"adoption":0.15,"quality":0.25,"ecosystem":0.1,"match_graph":0.45,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.550Z","last_scraped_at":"2026-05-03T13:59:55.151Z","last_commit":"2025-11-29T21:19:09Z"},"community":{"stars":81,"forks":11,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=jparkerweb--ai-assistant-prompts","compare_url":"https://unfragile.ai/compare?artifact=jparkerweb--ai-assistant-prompts"}},"signature":"bRk3vokmltuM0+LEgPyKFrijbkI8k9UR/gb7t3YvqUxdrZ8A/Hzd0hFf9OyO0E/TgK4oD97x9EZFdewl11eGCA==","signedAt":"2026-06-21T14:30:04.255Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/jparkerweb--ai-assistant-prompts","artifact":"https://unfragile.ai/jparkerweb--ai-assistant-prompts","verify":"https://unfragile.ai/api/v1/verify?slug=jparkerweb--ai-assistant-prompts","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"}}