The proposal One team. One workflow.
One team. One workflow.
One memory loop.
Decision Memory is in an early design-partner phase. Instead of a plan grid, we run a focused pilot with one team and one decision domain — so value is provable before anything scales.
Design-partner pilot
Free during the pilot.
- 1 team · 1 agent workflow · 1 decision domain
- 2–4 weeks, boundaries agreed up front
- Evidence intake & candidate decisions
- Human review, approval, durable records
- Search & Ask DM · conflict checks
- MCP access · search / context / conflicts / propose
Team
Contact us — planned.
- Multiple teams / decision domains
- Task-specific memory packages
- Structured agent write-back & routing
- Applied decision / rule references
Enterprise
Custom — planned.
- Full Rule / Skill registry
- Multi-object conflict detection
- Cross-agent rendering & sync
- Deployment & data-boundary requirements
Machine-readable pricing lives at /pricing.md.
The pilot
Six steps. One decision domain.
01
Select domain
02
Import evidence
03
Approve memory
04
Connect agent
05
Run tasks
06
Measure
What we measure together
Provable value in 2–4 weeks.
1 team · 1 decision domain Boundaries agreed up front 2–4 weeks
We measure
Decision recall
Can the team answer “why?” faster?
We measure
Context coverage
Did tasks get the right memory?
We measure
Conflict capture
What got caught before merge?
Baseline and target values are agreed with you before the pilot starts. No numbers are asserted before they're measured in your environment.