rp-why: Three Dimensions of AI Collaboration

Overview

The rp-why skill is a self-reflection framework that helps AI practitioners measure and improve their collaboration practice. It tracks three dimensions:

• DOK (Depth of Knowledge) - Cognitive complexity of human prompts, scored 1.0-4.0 using Webb's framework. "Adjusted" accounts for compression of intent.
• TM (Tool Maturity) - Intentional orchestration of AI tools, tracked through the Orchestra Model (Tiers 1-6).
• ADT (Agentic Delegation Trust) - The relationship between tool sophistication and cognitive depth. Derived from the TM x DOK matrix.

The intersection of these dimensions produces diagnostic zones that reveal the health of the collaboration practice.

How to Use This Skill

Installation

npx skills add https://github.com/block/agent-skills --skill rp-why

Available Commands

Alternative: Natural Language

You can also ask naturally:

You: Analyze my AI collaboration patterns
You: What's my DOK distribution for this session?
You: How does this session compare to my baseline?
You: Give me the full rp-why longitudinal report

When to Use

• First time: Run /rp-why init (or /rp-why baseline) to establish your starting point
• End of session: Run /rp-why current to reflect on your work
• Weekly: Run /rp-why compare to track progress against baseline
• Monthly: Run /rp-why overall for the full growth picture

The Three Dimensions

Dimension 1: DOK (Depth of Knowledge)

Measures the cognitive complexity of human prompts. Scored 1.0-4.0 using Webb's Depth of Knowledge framework.

Adjusted DOK accounts for compression of intent - when short prompts carry complex meaning due to established context. A compressed "proceed" that triggers a multi-step architectural deployment is not DOK 1.

DOK 3+4 % is the primary growth signal. It measures what proportion of your work operates at strategic or extended thinking levels.

Compression % tracks how often short directives carry complex intent. Compression emerges as trust deepens between practitioner and agent.

Dimension 2: TM (Tool Maturity) - Orchestra Model

Measures intentional orchestration - how deliberately you coordinate AI tools, agents, and workflows. Only counts actions the user deliberately initiated.

Dimension 3: ADT (Agentic Delegation Trust) - Diagnostic Zones

Measures the gap between Tool Complexity and Human Cognitive Depth. The TM x DOK matrix produces six zones:

Diagnostic Zone Matrix

              DOK 1         DOK 2          DOK 3           DOK 4
            (Recall)    (Application)  (Strategic)     (Extended)
           +-----------+--------------+-------------+-------------+
Tier 5-6   | Overpowered| Underutil-  |  Frontier   |  Frontier   |
(Symphony/ |            |   izing     |             |             |
 Virtuoso) |            |             |             |             |
           +-----------+--------------+-------------+-------------+
Tier 3-4   | Overpowered|  Expected   |   Growing   |  Frontier   |
(Ensemble/ |            |             |             |             |
 Chamber)  |            |             |             |             |
           +-----------+--------------+-------------+-------------+
Tier 1-2   |  Expected  |   Growing   |  Thinking   |  Thinking   |
(Solo/     |            |             |   Ahead     |   Ahead     |
 Duet)     |            |             |             |             |
           +-----------+--------------+-------------+-------------+

Report Formats

/rp-why init or /rp-why baseline

Establishes the starting point. Analyzes all available session history.

Report structure:

+==================================================================+
|                    rp-why . BASELINE                             |
+==================================================================+

DATA SUMMARY
--------------------------------------------------------------------
Period:           [start] - [end] ([N] days)
Sessions:         [N]
Prompts:          [N] classified

THREE DIMENSIONS
--------------------------------------------------------------------
DOK (Depth of Knowledge)     [score]
TM  (Tool Maturity)          Tier [N] . [Name]
ADT (Delegation Trust)       [Zone]

DIAGNOSTIC ZONE: [Zone Name]
--------------------------------------------------------------------
[Zone description from the six-zone model]

DOK DISTRIBUTION
--------------------------------------------------------------------
DOK 1 (Recall):       [bar]  [%]
DOK 2 (Application):  [bar]  [%]
DOK 3 (Strategic):    [bar]  [%]
DOK 4 (Extended):     [bar]  [%]

DOK 3+4:  [%]
Compression:  [%]

GROWTH TARGETS
--------------------------------------------------------------------
DOK 3+4 target:       [%]
Compression target:   [%]
Next TM tier:         Tier [N] . [Name] ([what it means])

--------------------------------------------------------------------
Baseline saved to: ~/.config/goose/rp-why-baseline.json
Run /rp-why current after sessions to track progress.

/rp-why current

Analyzes the active session. The quick-check mirror.

Report structure:

+==================================================================+
|                    rp-why . CURRENT SESSION                      |
+==================================================================+

SESSION SNAPSHOT
--------------------------------------------------------------------
Date:             [date]
Conversations:    [N]
Prompts:          [N] classified

THREE DIMENSIONS
--------------------------------------------------------------------
DOK (Adjusted)       [score]
TM  (Tool Maturity)  Tier [N] . [Name]
ADT (Delegation)     [Zone]

DIAGNOSTIC ZONE: [Zone Name]
--------------------------------------------------------------------
[Zone description]

DOK DISTRIBUTION
--------------------------------------------------------------------
DOK 1 (Recall):       [bar]  [%]
DOK 2 (Application):  [bar]  [%]
DOK 3 (Strategic):    [bar]  [%]
DOK 4 (Extended):     [bar]  [%]

DOK 3+4:  [%]     Compression:  [%]

PEAK MOMENT
--------------------------------------------------------------------
"[highest DOK prompt text, truncated]"  - DOK [N] . [Level Name]

GROWTH NUDGE
--------------------------------------------------------------------
[Contextual nudge based on diagnostic zone, not generic]

> [Reflection question tailored to current zone]

/rp-why compare

Delta report. Shows movement from baseline to now.

Report structure:

+==================================================================+
|                    rp-why . COMPARE                              |
+==================================================================+

COMPARING: Today ([date]) vs. Baseline ([baseline period])

THREE DIMENSIONS                    Baseline        Today        Delta
--------------------------------------------------------------------
DOK (Adjusted)                      [score]         [score]    [+/-%]
TM  (Tool Maturity)                 Tier [N]        Tier [N]   [+/-N]
ADT (Delegation Trust)              [Zone]          [Zone]     [arrow]

DIAGNOSTIC ZONE: [Baseline Zone] -> [Current Zone]
--------------------------------------------------------------------

DOK DISTRIBUTION                    Baseline        Today        Delta
--------------------------------------------------------------------
DOK 1 (Recall)                      [%]             [%]       [+/-pp]
DOK 2 (Application)                 [%]             [%]       [+/-pp]
DOK 3 (Strategic)                   [%]             [%]       [+/-pp]
DOK 4 (Extended)                    [%]             [%]       [+/-pp]

DOK 3+4                             [%]             [%]       [+/-pp]
Compression                         [%]             [%]       [+/-pp or "emerged"]

TRAJECTORY
--------------------------------------------------------------------
Direction:  [arrow] [Improving/Stable/Declining]
Signal:     [1-2 sentence interpretation of what the delta means]

WHAT SHIFTED
--------------------------------------------------------------------
* [Bullet interpretations of the most meaningful changes]
* [Focus on what the numbers mean for the collaboration practice]
* [Connect to diagnostic zone movement]

--------------------------------------------------------------------
Run /rp-why overall for full longitudinal analysis.

/rp-why overall

Full longitudinal report. The complete growth picture.

Report structure:

+==================================================================+
|                    rp-why . OVERALL                              |
+==================================================================+

FULL DATASET: [start] - [end] ([N] days)
Sessions: [N]  |  Prompts: [N]  |  Conversations: [N]

CURRENT STANDINGS
--------------------------------------------------------------------
DOK (Adjusted, full mean)    [score]
TM  (Tool Maturity)          Tier [N] . [Name]
ADT (Delegation Trust)       [Zone]

DOK 3+4:  [%]     Compression:  [%]     Floor:  [score]

ROLLING AVERAGE (Last 10 Sessions vs. First 10)
--------------------------------------------------------------------
                             First 10        Last 10          Delta
Adjusted DOK                 [score]         [score]        [+/-%]
DOK 3+4 %                   [%]             [%]           [+/-pp]
Compression %                [%]             [%]           [emerged/+/-pp]

PHASE ANALYSIS
--------------------------------------------------------------------
Phase         Dates              Sessions  DOK    DOK3+4  Comp   TM
-----         -----              --------  ---    ------  ----   --
[Auto-detected phases based on DOK trajectory shifts]

TRAJECTORY
--------------------------------------------------------------------
Peak DOK:     [score] ([date]  - [context])
Floor:        [score] ([interpretation])
Direction:    [arrow] [Narrative of the growth arc]

DOK DISTRIBUTION (Full Dataset)
--------------------------------------------------------------------
DOK 1 (Recall):       [bar]  [%]
DOK 2 (Application):  [bar]  [%]
DOK 3 (Strategic):    [bar]  [%]
DOK 4 (Extended):     [bar]  [%]

GROWTH STORY
--------------------------------------------------------------------
[2-3 sentence narrative interpreting the full arc. Not prescriptive.
Describes what the data reveals about how the collaboration practice
has evolved.]

--------------------------------------------------------------------
Data source: ~/.local/share/goose/sessions/sessions.db
Methodology: DOK keyword classification, compression detection
(short prompt + high response ratio + established context),
adjusted DOK (+1 level for compressed prompts, capped at 4).

Backward Compatibility

Reading Previous Baselines

If an existing baseline file uses the v3 format (Gas Town stages, quadrant terminology), the skill translates automatically:

Gas Town to Orchestra Mapping

For users familiar with the Gas Town stages:

Metrics Reference

Growth Nudges by Zone

Frontier

• "Operating at the productive edge. Document what works for others."
• "The collaboration is matched. Look for opportunities to extend into new domains."

Growing

• "Approaching a match. Keep pushing DOK 3+ work and the zone will shift."
• "Consider: what's one workflow you could delegate more fully?"

Expected

• "Healthy starting position. Growth comes from asking 'why' before implementing."
• "Try framing one task as a design decision rather than an execution request."

Thinking Ahead

• "Cognitive depth exceeds tool sophistication. Time to adopt more powerful orchestration."
• "Your thinking is ready for Tier [N+1]. Explore sub-agents or multi-step delegation."

Underutilizing

• "Powerful tools deserve powerful questions. Before each prompt: can this be more strategic?"
• "Batch simple queries. Reserve the agent for work that requires reasoning."

Overpowered

• "Significant mismatch. Consider whether this task needs an autonomous agent."
• "Opportunity: redirect this tool toward a problem that requires analysis or design."

Attribution

• Depth of Knowledge (DOK): Norman Webb (1997). Webb, N. L. Criteria for alignment of expectations and assessments in mathematics and science education.
• Orchestra Model (TM): Dakota Fabro (2026). Measuring Cognitive Complexity in Human-AI Collaboration.
• Three Dimensions Framework: Dakota Fabro (2026). rp-why longitudinal dataset, Block Builder Fellowship.
• Gas Town Stages (v1-v3 foundation): Steve Yegge, "Welcome to Gas Town" (January 2026).

Version History