The Expectation Setter recipe analyzes incoming tasks, communicates scope, complexity, deliverables, and limitations upfront before work begins. It automatically initializes progress tracking for complex tasks, preventing miscommunication and building trust through transparency. This recipe transforms ambiguous requests into clear, shared understanding between user and AI.
Recipe Name: RCP-001-001-012-EXPECT-SET – Proactive Expectation Management Recipe
RCP-001-001-012-EXPECT-SET
Communicates upfront what will happen, estimated complexity,likely duration, and known limitations before starting tasks.Automatically initializes PROGRESS_TRACKER recipe for ongoingmonitoring throughout task execution. Includes policy pre-checkto identify potential conflicts before work begins.
Requirements
How To Start
A Note From The Author of CRAFT
- After hundreds (perhaps thousands) of hours of using these recipes, I rarely need to use any of the CORE Cookbook recipes aside from Recipes RCP-001-001-002-HANDOFF-SNAPSHOT and RCP-001-001-002-HANDOFF-SNAPSHOT, but when I do, they are essential to the functioning of CRAFT. Also, the A.I. reads all of these recipes at the start of each session. This happens quietly in the background. Even though you may never need to call the recipe, the A.I. will know all of them and it helps the A.I. to understand what CRAFT is and how it works. Even if you rarely need to use these recipes, they are still working for you and are essential to the CRAFT Framework.
STEP 0: Policy Pre-Check
- Scan the task for sensitive categories before beginning any work. SENSITIVE CATEGORIES TO CHECK Platform capabilities and limitations Security or vulnerability research Personal data handling requirements Political or controversial topics Legal or medical advice requests IF potential conflict detected: #AI->H::PolicyCaution: (This task may trigger usage policies) #AI->H::RecommendedChange: (Consider focusing on [specific safe aspect])
STEP 1: Analyze Task Complexity
- Evaluate the requested task for complexity factors. COMPLEXITY FACTORS TO ASSESS Number of distinct steps required Dependencies on external resources Platform limitations involved Potential failure points Required user inputs Ambiguity level in request COMPLEXITY RATINGS SIMPLE: 1-2 steps, clear path MODERATE: 3-5 steps, some complexity COMPLEX: 6+ steps, multiple paths VERY COMPLEX: Many dependencies
STEP 2: Check Capabilities and Limitations
- Identify what can and cannot be done upfront. CAN DO List what AI can accomplish for this task. CANNOT DO List platform limitations that apply. List policy restrictions that apply. WILL SIMULATE List what will be demonstrated or mocked. REQUIRES USER ACTION List what user must provide or do.
STEP 3: Estimate Time and Effort
- Provide realistic estimates based on detail level. FOR MINIMAL DETAIL Provide quick estimate only. FOR BALANCED DETAIL Estimated Effort: Setup: approximately X exchanges Execution: approximately Y exchanges Total: approximately Z exchanges FOR COMPREHENSIVE DETAIL Detailed Breakdown: Phase 1: description (X exchanges) Phase 2: description (Y exchanges) Potential delays: what might extend timeline
STEP 4: Present Expectations Clearly
- Task Overview: [task_description] Complexity: [complexity level] What I Will Do: Concrete deliverable 1 Concrete deliverable 2 Concrete deliverable 3 What I Cannot Do: Limitation 1 with alternative workaround Limitation 2 with alternative workaround What You Will Need To Do: User action 1 User action 2 Estimated Exchanges: [total_exchanges] Potential Challenges: [main risk or complexity]
STEP 5: Initialize Progress Tracking
- If auto_init_progress is True: Extract steps from task analysis: 1. First major step 2. Second major step 3. Third major step #AI->H::Status: (Initializing progress tracking) Call PROGRESS_TRACKER with: task_name: [task_description] mode: initialize steps_list: [extracted steps] Show integrated view: Progress Tracking: Active Current Status: 0 percent – Ready to begin First Step: [first step description]
STEP 6: Request Confirmation
- #AI->H::Question: (Does this match your expectations? Should I proceed with this approach?) If user wants changes: #AI->H::RequestingFeedback: (What aspects should I adjust?) If user confirms: #AI->H::Status: (Beginning task with progress tracking active) Proceed to first step.
STEP 7: Adjust Expectations During Execution
- If scope changes during execution: #AI->H::Note: (Task complexity has increased due to [reason]) #AI->H::Status: (Updating expectations:) Original estimate: X exchanges Revised estimate: Y exchanges Reason: [specific discovery] Update PROGRESS_TRACKER accordingly.COMMON PATTERNS Pattern 1: Simple Information Request Example: What is X? Complexity: SIMPLE Deliverable: Clear explanation of X Exchanges: approximately 1 Pattern 2: Multi-Step Process Example: Create a complete Y Complexity: MODERATE Deliverables: Planning, drafting, refining Exchanges: approximately 5-7 Pattern 3: Complex Integration Example: Implement Z with constraints A, B, C Complexity: VERY COMPLEX Deliverables: Analysis, design, implementation, testing Exchanges: approximately 10-15 Challenges: Constraint conflicts, platform limits
How AI Reads This Recipe
When processing this recipe, the AI assistant: 1. First scans the task for policy-sensitive categories and communicates any concerns before proceeding with analysis. 2. Receives the task description and analyzes it for complexity factors and dependencies. 3. Checks capabilities against task requirements to identify what can and cannot be done. 4. Generates time and effort estimates based on the specified detail level parameter. 5. Presents clear expectations including what will be delivered, limitations, and user actions. 6. Initializes PROGRESS_TRACKER recipe if enabled to provide ongoing monitoring. 7. Requests confirmation before beginning work and adjusts expectations if user provides feedback. 8. Updates expectations during execution when scope or complexity changes.PROCESSING PRIORITY Priority Description ——– ———– CRITICAL Run policy pre-check first CRITICAL Analyze task before starting HIGH Identify limitations upfront MEDIUM Provide realistic estimates LOW Integrate with progress tracking
When to Use This Recipe
IDEAL USE CASES Before starting any multi-step or complex task to align expectations with user. When task involves platform limitations that should be communicated upfront. When accurate effort estimation helps user plan their time and involvement. When transparency about capabilities builds trust and prevents disappointment. When task may touch sensitive categories that require policy awareness.WHEN NOT TO USE THIS RECIPE For simple one-exchange information requests that do not require upfront planning. When user has explicitly requested immediate action without preliminary overview. When previous conversation has already established clear expectations for current task.
Recipe FAQ
Q: When should I use comprehensive detail level?
A: Use comprehensive for complex multi-phase projects
where detailed breakdown helps user understand full
scope and timeline expectations. Q: Can I skip progress tracking integration?
A: Yes, set auto_init_progress to False if you prefer
to manage progress updates manually or if the task
is too simple to warrant tracking. Q: What if my estimates turn out wrong?
A: The recipe includes Step 7 for adjusting expectations
during execution. Transparency about changes builds
trust even when estimates shift. Q: Does this work for very simple requests?
A: For simple requests like explaining a concept, the
recipe automatically provides minimal expectations
without overwhelming the user. Q: How does this integrate with other recipes?
A: EXPECT-SET initializes PROGRESS_TRACKER automatically
and works with ERROR-TRANS for communicating any
limitations discovered during analysis. Q: What does STEP 0 Policy Pre-Check do?
A: It scans the task for sensitive categories like
security research, personal data, or political
topics that might trigger platform policies,
allowing proactive communication before work begins.
A: Use comprehensive for complex multi-phase projects
where detailed breakdown helps user understand full
scope and timeline expectations. Q: Can I skip progress tracking integration?
A: Yes, set auto_init_progress to False if you prefer
to manage progress updates manually or if the task
is too simple to warrant tracking. Q: What if my estimates turn out wrong?
A: The recipe includes Step 7 for adjusting expectations
during execution. Transparency about changes builds
trust even when estimates shift. Q: Does this work for very simple requests?
A: For simple requests like explaining a concept, the
recipe automatically provides minimal expectations
without overwhelming the user. Q: How does this integrate with other recipes?
A: EXPECT-SET initializes PROGRESS_TRACKER automatically
and works with ERROR-TRANS for communicating any
limitations discovered during analysis. Q: What does STEP 0 Policy Pre-Check do?
A: It scans the task for sensitive categories like
security research, personal data, or political
topics that might trigger platform policies,
allowing proactive communication before work begins.
Actual Recipe Code
(Copy This Plaintext Code To Use)
===========================================================# START RECIPE-ID: RCP-001-001-012-EXPECT-SET-v2.00c# ===========================================================EXPECTATION_SETTER_RECIPE = Recipe( recipe_id="RCP-001-001-012-EXPECT-SET-v2.00c", title="Proactive Expectation Management with Progress Tracker Integration", description="Communicates upfront what will happen, estimated complexity, likely duration, and known limitations before starting tasks, automatically initializes PROGRESS_TRACKER for monitoring", category="CAT-001", subcategory="SUBCAT-004-Communication-Flow", difficulty="medium", parameters={ "task_description": { "type": "string", "required": True, "description": "Task to set expectations for" }, "auto_init_progress": { "type": "boolean", "required": False, "default": True, "description": "Auto-init PROGRESS_TRACKER" }, "detail_level": { "type": "string", "required": False, "default": "balanced", "options": ["minimal", "balanced", "comprehensive"], "description": "Detail level for expectations" } }, prompt_template=""" #H->AI::Directive: (Set clear expectations before beginning [task_description]) #H->AI::Context: (Prevent overpromising, integrate with progress tracking) STEP 0: POLICY PRE-CHECK ======================== Scan task for sensitive categories: Platform capabilities/limitations Security/vulnerability research Personal data handling Political topics IF potential conflict detected: #AI->H::PolicyCaution: (This task may trigger usage policies) #AI->H::RecommendedChange: (Consider focusing on [specific safe aspect]) STEP 1: TASK ANALYSIS ===================== Analyze the requested task for: COMPLEXITY FACTORS: Number of distinct steps required Dependencies on external resources Platform limitations involved Potential failure points Required user inputs Ambiguity level COMPLEXITY RATING: SIMPLE (1-2 steps, clear path) MODERATE (3-5 steps, some complexity) COMPLEX (6+ steps, multiple paths) VERY COMPLEX (many dependencies) STEP 2: CAPABILITY CHECK ======================== Identify any limitations upfront: CAN DO: [List what AI can accomplish] CANNOT DO: [Platform limitations] [Policy restrictions] WILL SIMULATE: [What will be demonstrated] REQUIRES USER ACTION: [What user must provide/do] STEP 3: TIME/EFFORT ESTIMATION ============================== Provide realistic estimates: For [detail_level] == minimal: Quick estimate only For [detail_level] == balanced: ESTIMATED EFFORT: Setup: approximately X exchanges Execution: approximately Y exchanges Total: approximately Z exchanges For [detail_level] == comprehensive: DETAILED BREAKDOWN: Phase 1: [description] (X exchanges) Phase 2: [description] (Y exchanges) Potential delays: [what might extend] STEP 4: SET EXPECTATIONS ======================== Task Overview: [task_description] Complexity: [complexity_level] What I Will Do: [Concrete deliverable 1] [Concrete deliverable 2] [Concrete deliverable 3] What I Cannot Do: [Limitation 1] – Alternative: [workaround] [Limitation 2] – Alternative: [workaround] What You Will Need To Do: [User action 1] [User action 2] Estimated Exchanges: approximately [total] Potential Challenges: [Main risk or complexity] STEP 5: PROGRESS TRACKER INTEGRATION ==================================== IF [auto_init_progress] == True: Extract steps from analysis: 1. [First major step] 2. [Second major step] 3. [Third major step] #AI->H::Status: (Initializing progress tracking) Call PROGRESS_TRACKER with: task_name: [task_description] mode: initialize steps_list: [extracted steps] Show integrated view: Progress Tracking: Active Current Status: 0 percent – Ready to begin First Step: [step_1_description] STEP 6: CONFIRMATION REQUEST ============================ #AI->H::Question: (Does this match your expectations? Should I proceed with this approach?) IF user wants changes: #AI->H::RequestingFeedback: (What aspects should I adjust?) IF user confirms: #AI->H::Status: (Beginning task with progress tracking active) Proceed to first step STEP 7: EXPECTATION ADJUSTMENTS =============================== During execution, if scope changes: #AI->H::Note: (Task complexity has increased due to [reason]) #AI->H::Status: (Updating expectations:) Original estimate: approximately X exchanges Revised estimate: approximately Y exchanges Reason: [specific discovery] Update PROGRESS_TRACKER accordingly COMMON PATTERNS: ================ Pattern 1: Simple Information Request What is X? Complexity: SIMPLE Deliverable: Clear explanation of X Exchanges: approximately 1 Pattern 2: Multi-Step Process Create a complete Y Complexity: MODERATE Deliverables: Planning, drafting, refining Exchanges: approximately 5-7 Pattern 3: Complex Integration Implement Z with constraints A, B, C Complexity: VERY COMPLEX Deliverables: Analysis, design, implementation, testing Exchanges: approximately 10-15 Challenges: Constraint conflicts, platform limits #H->AI::OnError: (If unable to estimate, provide ranges with uncertainty levels) """)# ===========================================================# END RECIPE-ID: RCP-001-001-012-EXPECT-SET-v2.00c# ===========================================================
