Companion standards

DD6 + CIRK: not competing, not overlapping.

DD6 classifies the problem (upstream). CIRK classifies execution (downstream). Together they provide a complete governance trail from intake to delivery.

DD6 CIRK
Question How much discovery does this problem need? How safely can this task be executed?
Stage Upstream (intake → spec) Downstream (spec → delivery)
Input Raw intake Defined spec or task
Output Discovery depth Execution policy
Dimensions 6 (I, D, S, T, P, B) 4 (C, I, R, K)
Score range 6–18 4–12

The pipeline

From problem to delivery.

Intake
DD6
Discovery
Spec
CIRK
Execution
Delivery

"How much discovery does the problem need?" → "How safely can this execute?"

Why it matters

Better discovery produces better execution.

CIRK scores are more accurate when the spec is well-defined. DD6 ensures that complex problems receive proportional discovery before reaching CIRK.

Without DD6

  • A vague problem skips discovery entirely
  • The spec is ambiguous or incomplete
  • CIRK scores are uncertain because the spec is unclear
  • Agent execution is riskier than the score suggests

With DD6

  • The problem is explored proportionally to its complexity
  • The spec that arrives at CIRK is well-defined
  • CIRK scores accurately reflect execution reality
  • Agent execution matches the governance policy

Feed-forward

DD6 output improves CIRK accuracy.

Some DD6 dimensions have downstream echoes in CIRK. These are tendencies, not deterministic mappings.

DD6 dimension CIRK echo Relationship
T Testability R Review Hard to test (T3) → higher human review effort (R)
D Domain C Context Deep domain (D3) → more context needed (C2–C3)
P Precedent I Iteration No precedent (P3) → more iteration cycles (I2–I3)
S Stakeholder R Review Multi-stakeholder (S3) → broader review (R2–R3)

A deep DD6 problem can produce tasks that are trivial in CIRK — that is the whole point of good discovery.

Adoption

DD6 stands alone.

Teams can adopt DD6 independently of CIRK. However, teams that also adopt CIRK will find that the governance trail is complete from problem to delivery, fewer execution failures occur, and scoring becomes more consistent.