Faster alignment
Shared notes reduce “what did we learn?” meetings after the fact.
ReadableShareable
Research Mode
Answer hard questions without mutating the repo
Research Mode is read-first exploration: trace behavior, compare approaches, and document what you learned - before any writes land.
- Read-first · no silent writes
- Great for spikes and incident prep
- Same session as delivery modes
research-notes.md - read-only output
Question“Why is this worker retrying forever?”
TraceCall path + config surfaces
CompareTwo viable fix directions
NotesHandoff for implementation
Research Mode
From unclear behavior to a written research brief
Use Research when you need confidence before touching production paths - especially during incidents, onboarding, and refactors.
- Explore codepaths and configuration without writing to disk
- Compare approaches with explicit tradeoffs
- Produce notes your team can implement from later
- Lower risk while incident or spike context is hot
- Switch into coding modes when you are ready to edit
Research goal:
Incident
Context
Depth
Output
research-notes.md
Preview
// Research question:
"Map retry behavior for the billing worker"
Observed behavior: - Retries on 5xx with exponential backoff - No cap on total attempts Hypotheses: - Missing maxAttempts in config - DLQ path never wired Recommended next step: - Confirm env flags in staging, then patch with cap + DLQ
1Ask
2Trace
3Compare
4Notes
5Handoff
What teams get from one Research pass
Lower blast radius
No repo mutations while you are still discovering the real failure mode.
Read-firstSafer spikes
Cleaner handoff to build
Implementation starts with a written map instead of tribal memory.
TraceabilityActionable
Best-fit scenarios
Use Research Mode before you commit to edits
Spikes, incident triage, and unfamiliar modules are common starting points.
Incidents
Triage before patchOnboarding
New service or monolith areaRefactors
Map blast radius firstAPIs
Contract and error semanticsCompliance
Evidence-friendly notes
Read-only
Lower risk
Faster clarity
Better handoffs
How it works
Run Research Mode in three moves
Keep exploration disciplined and documented.
Frame the question
State what you need to know and what evidence would convince you.
Trace read-only
Follow call paths, configs, and data flow without writes.
Publish notes
Capture conclusions, risks, and recommended next actions.
FAQ
Research Mode FAQ
Does Research write files?
No. It is designed for read-first exploration; switch to your normal coding flow when you are ready to edit.
Is this just “ask ChatGPT”?
No. The goal is repository-grounded notes you can attach to tickets, incidents, or PRs.
When should we use Research vs Strategy?
Research clarifies behavior and unknowns; Strategy resolves directional bets with owners and milestones.
How do we measure value?
Track time-to-root-cause on incidents, spike duration, and rework after handoff to implementation.
Pricing
Pricing
Use Research Mode within the same Kodus plans and usage model.
Best to Start
Starter
For individual usage.
$20/mo
- 10M tokens / month
- 300 iterations / month
- Standard model routing
- Chat history + projects
- Community support
Team
For small teams.
$100/mo
- 70M tokens / month
- 2,500 iterations / month
- Full routing + Review + Strategy
- Bring your own local model
- Teams (up to 2 members)
- Priority support
- Audit log access
Scale
For larger organizations.
$200/mo
- 300M tokens / month
- 7,500 iterations / month
- Unlimited team members
- All models + custom routing
- Dedicated support channel
- Early access to beta features
- No annual contract
- Tokens reset monthly
- Switch plans anytime
Have invite code? Get Access Now