Assessment output structure¶
Each run writes assessments/YYYY-MM-DD-assessment.md with the following
sections, in order. Every section is filled with specific evidence —
paths, counts, dates.
Header¶
The report leads with two lines — the current level and the next step — so the headline answer is legible at a glance. Coherence (the Habitat/Workflow Gap) and the cognitive read follow as secondary fields.
**AI Readiness — Habitat Maturity**: Level N (Verb)
**Next Step / Gap**: +X to Level N+1 (NextVerb)
**Habitat/Workflow Gap**: <signed gap> (<regime>) (coherence; secondary)
**Assessed level**: Level N — <Level Name> (the cognitive read)
**Date**: YYYY-MM-DD
The level verb is the model's Agent-behaviour archetype: L1
Dictating, L2 Commanding, L3 Regulating, L4 Orchestrating, L5
Supervising. Next Step / Gap is (N+1) − the 14-dimension mean — the
distance to the next level (at L5: "at the top — sustaining"). It is
distinct from the Habitat/Workflow Gap, which is the
coherence diagnostic
(cognitive − operational) and can differ in both value and sign.
Habitat Document Discovery¶
A table of the habitat documents found — instruction files, constraint documents, specs, reflection/decision records — each with its path and the content markers that confirmed the match. Produced before any maturity claim. If two files plausibly fill the same role, the assessment stops and asks which is canonical rather than guessing.
Observable Evidence¶
Every signal found (with path) and every signal not found. The absences matter as much as the presences.
Clarifying Responses¶
The 3–5 questions asked (one at a time) and the answers given — the behavioural detail the filesystem can't supply.
AI Readiness Score — five readiness dimensions¶
Output 1. A public-facing breakdown across five readiness dimensions — Context · Conventions · Architectural guidance · Guardrails · Agent readiness — each placed L1–L5, mapped from the same evidence as the rest of the report. It is a view, not a separate scoring model: the headline level remains the fourteen-dimension Habitat Maturity. (Mapping: Context ← Context Engineering / Teams provide; Conventions ← HARNESS Conventions / convention files; Architectural guidance ← Architectural Constraints / specs; Guardrails ← Guardrail Design / Testing / Observability; Agent readiness ← Composition / Workflow / Agents.)
Habitat Maturity Profile¶
All fourteen model dimensions placed L1–L5, each reported with the model's verb (e.g. Testing: L3 — Verifying). Behavioural dimensions placed without direct evidence are tagged (inferred). Ends with the headline Habitat Maturity Level (the rounded mean, weakest dimensions named).
Level Assessment¶
The cognitive level (L0–L5), its name, and a one-line rationale anchored in the weakest discipline.
Discipline Maturity¶
| Discipline | Strength (0–5) | Evidence |
|---|---|---|
| Context Engineering | N | … |
| Architectural Constraints | N | … |
| Guardrail Design | N | … |
Operational Axes (Part D)¶
The four discipline-aligned headline dimensions — Composition, Testing, Observability, Governance — lifted from the profile as a focused view.
Habitat/Workflow Gap¶
The signed gap and its regime. See The Habitat/Workflow Gap.
Strengths · Gaps¶
Top three each, anchored in evidence.
Prioritised Improvement Plan¶
Output 2. A ranked list of what to build first to reach the next level, ordered by what the team needs to develop and what the organisation needs to provide (the latter maps to the model's Teams provide dimension). Each item ties to a readiness dimension or discipline gap.
Next Steps¶
The generated explanation of what to do about the gap, naming one TechTalk engagement matched to it — deliberately one recommendation, not a menu. A menu reads like marketing; a specific recommendation reads like advice. It closes with a fixed paragraph introducing the purpose of a review. This section carries no call to action of its own.
Assessment Review¶
The report's single call to action — a one-hour session reviewing the report with the assessment's author, framed as the continuation of the assessment rather than as consulting promotion: why these scores came out this way, which findings matter, where to invest first, what can wait, and how to explain the results internally. Fixed copy, identical in every report.
Reading Path¶
The secondary, self-guided alternative — the specific chapter of The Sovereign Engineer that closes the weakest discipline gap, gap-anchored rather than a generic plug, with the Leanpub link.
A shareable HTML version is offered on request.