Task Context Skill

This skill loads the context file for the current task, providing focused information for the executor or validator session.

Trigger

This skill is automatically invoked when a Professor Frink session starts, loading the appropriate context file.

Usage

/task-context [task_id]

If no task_id is provided, uses the current task from .frink/state.json.

Behavior

1. •

   Identify Current Task

   • •Read task ID from argument or state file
   • •Locate context file: .frink/context/task-{id}-context.md
2. •

   Load Context

   • •Read the task context file
   • •Display task details, acceptance criteria, and relevant standards
3. •

   Load Additional Context

   • •Check for HITL feedback: .frink/HITL_FEEDBACK.md
   • •Check for progress notes: .frink/progress.txt
4. •

   Display Summary

   code

   Task Context Loaded
   
   Task: 2.1 Create Terraform AWS provider configuration
   Group: Infrastructure - Terraform (1 of 12)
   
   Acceptance Criteria:
   - [ ] Provider configuration for AWS
   - [ ] Backend configuration for state
   - [ ] Variables file for environment config
   
   Relevant Standards: 2 sections loaded
   HITL Feedback: None
   Progress Notes: 5 entries
   

Context File Structure

Each task context file contains:

• •Task details (ID, title, group)
• •Acceptance criteria
• •Implementation notes
• •Files to create/modify
• •Verification commands
• •Relevant standards excerpts
• •Spec requirements excerpts
• •Mission context
• •Technology notes

Example

# Load context for current task
/task-context

# Load context for specific task
/task-context 2.1

Integration

This skill is used by:

• •agents/task-executor.md - To understand what to implement
• •agents/task-validator.md - To know what to validate
• •agents/task-fixer.md - To understand project standards

The context files are generated during /professor-frink:init by the context-builder agent.