Install any skill in seconds. Free to start, no credit card required.
Get Started Free →DACI decision facilitation framework (Driver, Approver, Contributor, Informed) for clarifying decision ownership, reducing decision thrash, role assignment, and governance design across product teams.
.claude/skills/borghei-daci-framework/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 161% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 44% | 0% |
Clarify decision ownership and reduce decision thrash using the DACI framework (Driver, Approver, Contributor, Informed). Unlike RACI which focuses on task responsibility, DACI is purpose-built for product decisions -- who drives the decision to closure, who has veto power, who provides input, and who needs to know.
The four roles in brief: Driver (exactly one, drives to closure), Approver (1-2 max, holds veto), Contributor (input without veto), Informed (notified, not consulted). Build a chart by mapping current state, finding pain points, designing target state, and rolling out a 30/60/90 transition. See the playbook reference for role rules, the 7-step build sequence, and health metrics.
Before building the DACI chart, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
In Scope: DACI chart creation, current-state mapping, target-state design, transition planning, governance health metrics, pain point identification, decision ownership clarity.
Out of Scope: Task assignment (use RACI), project execution tracking (use sprint planning), individual performance management, organizational design beyond decision governance.
Important Caveats: DACI works best when leadership commits to respecting the framework. Without executive buy-in, Drivers may lack the authority to actually drive decisions. Start with 3-5 high-impact decisions rather than trying to map everything at once.
| Integration | Direction | What Flows | |---|---|---| | create-prd/ | Feeds into | DACI decisions inform PRD Contacts section and decision log | | identify-assumptions/ | Complements | Surfaces assumptions about who has authority | | brainstorm-okrs/ | Complements | OKR ownership aligns with DACI decision ownership | | summarize-meeting/ | Feeds into | Meeting summaries capture DACI decision outcomes | | senior-pm/ | Complements | Portfolio-level DACI for cross-project decisions |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 25,187 | 30,605 | +22% | 1 | 1 | 0% | 3,966 | 5,729 | +44% | 0 | 0 | — |
case-02 | fail→fail | 20,298 | 19,230 | -5% | 1 | 1 | 0% | 3,200 | 3,929 | +23% | 0 | 0 | — |
case-03 | fail→fail | 25,253 | 29,880 | +18% | 1 | 1 | 0% | 4,036 | 5,570 | +38% | 0 | 0 | — |
case-04 | fail→pass | 8,981 | 19,420 | +116% | 1 | 1 | 0% | 1,541 | 4,016 | +161% | 0 | 0 | — |
case-05 | fail→pass | 12,347 | 11,602 | -6% | 1 | 1 | 0% | 1,829 | 2,716 | +48% | 0 | 0 | — |
case-06 | pass→pass | 13,065 | 14,055 | +8% | 1 | 1 | 0% | 2,005 | 3,110 | +55% | 0 | 0 | — |
case-07 | pass→pass | 13,237 | 13,590 | +3% | 1 | 1 | 0% | 2,093 | 3,078 | +47% | 0 | 0 | — |
case-08 | pass→pass | 13,676 | 14,571 | +7% | 1 | 1 | 0% | 2,061 | 3,077 | +49% | 0 | 0 | — |
case-09 | fail→pass | 18,521 | 14,229 | -23% | 1 | 1 | 0% | 2,960 | 3,110 | +5% | 0 | 0 | — |
case-10 | pass→pass | 12,155 | 16,772 | +38% | 1 | 1 | 0% | 2,313 | 3,808 | +65% | 0 | 0 | — |
case-11 | fail→pass | 13,111 | 9,308 | -29% | 1 | 1 | 0% | 2,144 | 2,455 | +15% | 0 | 0 | — |
case-12 | fail→fail | 20,216 | 13,804 | -32% | 1 | 1 | 0% | 3,211 | 3,140 | -2% | 0 | 0 | — |
case-13 | pass→pass | 8,331 | 8,523 | +2% | 1 | 1 | 0% | 1,260 | 2,222 | +76% | 0 | 0 | — |
case-14 | fail→pass | 19,330 | 21,889 | +13% | 1 | 1 | 0% | 2,880 | 4,159 | +44% | 0 | 0 | — |
case-15 | fail→pass | 10,902 | 15,063 | +38% | 1 | 1 | 0% | 1,693 | 3,188 | +88% | 0 | 0 | — |
case-16 | fail→fail | 16,266 | 15,209 | -6% | 1 | 1 | 0% | 2,389 | 3,364 | +41% | 0 | 0 | — |
case-17 | pass→pass | 14,831 | 16,141 | +9% | 1 | 1 | 0% | 2,357 | 3,354 | +42% | 0 | 0 | — |
case-18 | pass→pass | 12,131 | 5,677 | -53% | 1 | 1 | 0% | 2,228 | 1,832 | -18% | 0 | 0 | — |
case-19 | pass→pass | 4,835 | 5,552 | +15% | 1 | 1 | 0% | 759 | 1,865 | +146% | 0 | 0 | — |
case-20 | pass→pass | 6,964 | 8,409 | +21% | 1 | 1 | 0% | 1,089 | 2,179 | +100% | 0 | 0 | — |
case-21 | fail→pass | 8,325 | 3,486 | -58% | 1 | 1 | 0% | 1,215 | 1,478 | +22% | 0 | 0 | — |
case-22 | fail→pass | 16,235 | 19,661 | +21% | 1 | 1 | 0% | 2,534 | 3,981 | +57% | 0 | 0 | — |
case-23 | pass→pass | 11,664 | 11,014 | -6% | 1 | 1 | 0% | 1,746 | 2,494 | +43% | 0 | 0 | — |
case-24 | pass→pass | 11,050 | 11,024 | -0% | 1 | 1 | 0% | 1,653 | 2,529 | +53% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +33 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.