Agent Type: Background Confidence AssessmentTools: Read, Glob, Grep, BashSpecial: Runs in background with isolated worktree
Overview
The Scout agent performs confidence-gated exploration to assess whether you have enough context to implement a task successfully. It scores readiness 0-100 across five dimensions and delivers a GO/HOLD verdict. Scout runs in the background, allowing you to continue working while it explores.When to Use
Use Scout before starting implementation of:Unfamiliar Code
Working in parts of the codebase you haven’t touched before
Complex Tasks
Tasks with unclear scope or many unknowns
High Risk Changes
Changes that could break existing functionality
New Features
Building features without clear patterns to follow
Configuration
Special Capabilities
Scout runs in the background, allowing you to continue working while it explores the codebase.
Scout runs in an isolated worktree to avoid interfering with your main work session.
Workflow
Scout follows a confidence-building workflow:Confidence Scoring
Scout evaluates five dimensions, each worth 0-20 points:1. Scope Clarity (0-20)
20 points: Crystal Clear
20 points: Crystal Clear
- Know exactly which files need changes
- Understand all modifications required
- Can enumerate every change needed
10-15 points: Mostly Clear
10-15 points: Mostly Clear
- Know most files that need changes
- Some uncertainty about edge cases
- Can estimate change scope
0-5 points: Unclear
0-5 points: Unclear
- Unsure which files to modify
- Don’t understand the full scope
- Many unknowns remain
2. Pattern Familiarity (0-20)
20 points: Clear Patterns
20 points: Clear Patterns
- Codebase has similar features to follow
- Patterns are consistent and documented
- Can copy existing approach
10-15 points: Some Patterns
10-15 points: Some Patterns
- Found some related code
- Patterns exist but inconsistent
- Need to adapt existing patterns
0-5 points: No Patterns
0-5 points: No Patterns
- No similar features exist
- First time implementing this type of feature
- Need to create new patterns
3. Dependency Awareness (0-20)
20 points: Full Awareness
20 points: Full Awareness
- Know all dependencies and dependents
- Understand impact of changes
- Can predict breaking changes
10-15 points: Partial Awareness
10-15 points: Partial Awareness
- Know major dependencies
- Some unknowns about impact
- Can identify most breaking changes
0-5 points: Unknown
0-5 points: Unknown
- Don’t know what depends on this code
- Can’t predict impact
- High risk of breaking things
4. Edge Case Coverage (0-20)
20 points: Comprehensive
20 points: Comprehensive
- Can enumerate all edge cases
- Know how to handle errors
- Understand validation requirements
10-15 points: Adequate
10-15 points: Adequate
- Know major edge cases
- Some uncertainty about errors
- Basic validation understood
0-5 points: Gaps
0-5 points: Gaps
- Don’t know edge cases
- Unsure how to handle errors
- Validation requirements unclear
5. Test Strategy (0-20)
20 points: Clear Strategy
20 points: Clear Strategy
- Know exactly how to test changes
- Test patterns exist to follow
- Can verify all functionality
10-15 points: Basic Strategy
10-15 points: Basic Strategy
- Have general testing approach
- Some test patterns available
- Can verify main functionality
0-5 points: No Strategy
0-5 points: No Strategy
- Don’t know how to test
- No test patterns to follow
- Unclear how to verify changes
Output Format
Example Scenarios
Rules and Constraints
Never Edit Files
Never Edit Files
Scout performs read-only exploration only. It never modifies the codebase.
Be Honest About Gaps
Be Honest About Gaps
A false GO wastes more time than a HOLD. Scout is brutally honest about confidence.
Re-Score After Context
Re-Score After Context
If score < 70, Scout gathers more context and re-scores. Maximum 2 rounds before escalating.
Isolated Execution
Isolated Execution
Runs in isolated worktree to avoid interfering with your main work session.
Best Practices
Comparison with Planner
| Feature | Scout | Planner |
|---|---|---|
| Output | GO/HOLD verdict | Detailed plan |
| Scoring | 0-100 confidence | Not applicable |
| Execution | Background | Foreground |
| Isolation | Worktree | None |
| Re-attempts | Auto (max 2) | Manual |
| Use case | Quick confidence check | Detailed planning |
Integration Example
Troubleshooting
Scout Takes Too Long
Scout Takes Too Long
Scout runs in background and should not block your work. If it seems slow, check:
- Very large codebase? May need more time
- Complex dependencies? Requires thorough exploration
- Wait for verdict or continue with other tasks
Always Getting HOLD
Always Getting HOLD
If Scout consistently returns HOLD:
- Task might be too complex for immediate implementation
- Codebase might lack patterns for this feature
- Consider breaking task into smaller pieces
- Use Planner for architectural guidance
GO But Still Confused
GO But Still Confused
If Scout says GO but you’re uncertain:
- Review the findings section carefully
- Score may be borderline (70-75)
- Run Planner for more detailed guidance
- Trust your judgment—human intuition matters
Next Steps
Orchestrator
Multi-phase implementation after GO verdict
Planner
Detailed planning for HOLD scenarios