Output Formats
Customize how ReasonKit displays results.
Available Formats
Pretty (Default)
Human-readable output with colors and formatting.
rk think "question" --format pretty
╔════════════════════════════════════════════════════════════╗
║ POWERCOMBO ANALYSIS ║
║ Profile: balanced ║
╚════════════════════════════════════════════════════════════╝
┌────────────────────────────────────────────────────────────┐
│ 💡 GIGATHINK: 10 Perspectives │
├────────────────────────────────────────────────────────────┤
│ 1. FINANCIAL: ... │
│ 2. CAREER: ... │
│ ... │
└────────────────────────────────────────────────────────────┘
JSON
Machine-readable JSON output for programmatic use.
rk think "question" --format json
{
"version": "0.1.0",
"timestamp": "2026-01-15T10:30:00Z",
"profile": "balanced",
"question": "Should I take this job?",
"results": {
"gigathink": {
"perspectives": [
{"category": "FINANCIAL", "content": "..."},
{"category": "CAREER", "content": "..."}
]
},
"laserlogic": {
"flaws": [
{"claim": "...", "issue": "...", "evidence": "..."}
]
},
"bedrock": {
"core_question": "...",
"first_principles": ["...", "..."],
"eighty_twenty": "..."
},
"proofguard": {
"verified": [...],
"unverified": [...],
"verdict": "..."
},
"brutalhonesty": {
"uncomfortable_truths": [...],
"questions": [...]
}
},
"synthesis": "...",
"metadata": {
"execution_time_ms": 12500,
"tokens_used": 4892,
"model": "claude-3-sonnet-20240229"
}
}
Markdown
Documentation-friendly markdown output.
rk think "question" --format markdown
# PowerCombo Analysis
**Question:** Should I take this job?
**Profile:** balanced
**Time:** 2026-01-15 10:30:00
## 💡 GigaThink: 10 Perspectives
1. **FINANCIAL**: ...
2. **CAREER**: ...
...
## ⚡ LaserLogic: Reasoning Check
### Flaw 1: "..."
- Issue: ...
- Evidence: ...
## 🪨 BedRock: First Principles
**Core Question:** ...
## 🛡️ ProofGuard: Verification
| Claim | Status | Evidence |
| ----- | ---------- | -------- |
| ... | ✓ Verified | ... |
## 🔥 BrutalHonesty: Reality Check
### Uncomfortable Truths
1. ...
2. ...
---
## Synthesis
...
Output Options
Summary Only
Get just the synthesis without full tool outputs:
rk think "question" --summary-only
Specific Tools Only
Output only specific tool results:
# Just BrutalHonesty
rk think "question" --tools brutalhonesty
# Multiple tools
rk think "question" --tools gigathink,laserlogic
Include/Exclude Metadata
# Include timing and token info
rk think "question" --show-metadata
# Exclude metadata
rk think "question" --no-metadata
Maximum Length
Limit output length:
rk think "question" --max-length 1000
Piping and Redirection
Save to File
# JSON for later processing
rk think "question" --format json > analysis.json
# Markdown for documentation
rk think "question" --format markdown > analysis.md
Pipe to Other Tools
# Extract specific field with jq
rk think "question" --format json | jq '.results.brutalhonesty.uncomfortable_truths'
# Count perspectives
rk think "question" --format json | jq '.results.gigathink.perspectives | length'
# Pretty print JSON
rk think "question" --format json | jq '.'
Chain with Other Commands
# Email results
rk think "question" --format markdown | mail -s "Analysis" user@example.com
# Add to notes
rk think "question" --format markdown >> ~/notes/decisions.md
# Render markdown
rk think "question" --format markdown | glow -
Configuration
Set default format in config:
[output]
format = "pretty" # pretty, json, markdown
color = true
show_timing = true
show_tokens = false
box_style = "rounded" # rounded, sharp, ascii
Custom Templates
Create custom output templates (MCP (Pro) feature):
[output.templates.minimal]
include_tools = ["gigathink", "brutalhonesty"]
include_synthesis = true
include_metadata = false
max_perspectives = 5
Use custom template:
rk think "question" --template minimal