Define Strategy

⛔ This skill MUST be read in full — not skimmed. Formal review gates depend on its workflow.
Skipping steps silently bypasses quality checks. Missing gates = undetected breakages.

Produces the Strategy artifact and Roadmap. Takes the S4 vision as input. Executed by the Team Strategist role.

Philosophy

Good strategy eliminates options. It forces hard trade-offs that drive specialization. If the output doesn't make you uncomfortable about what you're NOT doing, it's not a strategy — it's a wishlist.

• Default to less. Doing fewer things well beats doing more things poorly.
• Every "yes" to an initiative is a "no" to something else. Make those nos explicit.
• Delegation is the first choice. Adding roles is the last.
• The best infrastructure is the infrastructure you don't build.

VSM S3 Alignment

S3 Function	Step	Ponytail Check
Bridge current to future	1: Gap analysis	Which gaps should we accept?
Balance autonomy with goals	2: Horizon mapping	What can be deferred? Best infra is unbuilt.
Resource allocation	3: Initiative generation	What's the one-line version?
System re-organization	4-5: Role gap + workflows	Delegate → skills → expand agreement → add role (last)
Performance management	6: Coherence review	What did we say no to?
Handoff to execution	7: Roadmap → Coordinator	—

What you are missing

• workflow/01-gap-analysis.md — current vs desired, competitor research, internal analytics, AI review
• workflow/02-horizon-mapping.md — Three Horizons, ponytail check, human gate
• workflow/03-initiative-generation.md — one-off + ongoing Actions, human gate
• workflow/04-role-gap-analysis.md — map initiatives to existing roles, AI review
• workflow/05-workflows-plan.md — delegate → skills → expand → add, human gate
• workflow/06-coherence-review.md — internal consistency, trade-off check
• workflow/07-roadmap.md — handoff to Coordinator, human gate

Memory System Integration (S9)

Before Strategy — Load Existing Strategy Points

Before starting gap analysis, query the epistemic graph for prior strategy claims. This surfaces assumptions, constraints, and decisions from previous strategy cycles — preventing drift and enabling cumulative reasoning.

# From the agent-infra checkout (scripts/ lives there — symlinked into consumers):
node scripts/tortoise-memory.mjs query-strategies

Also query for relevant decisions and assumptions:

node scripts/tortoise-memory.mjs query-prior-research --domain "<strategy-domain>"

Interpretation:

• Existing strategy Points found: Summarize prior decisions, constraints, and assumptions. Use them as inputs to the gap analysis — what changed since the last strategy? What assumptions held true? Which didn't?
• status: "not_configured" (no usable TORTOISE_API_KEY — never set up, or the key was rejected with HTTP 401/403) or status: "tortoise_unavailable" (the store was unreachable): skip — memory is not available here. Proceed without prior context.
• Zero results: First strategy cycle for this domain. Note "no prior strategy Points found."

After Strategy — Write Strategy Points to Memory

After the Roadmap step completes, log the strategy's key claims to the epistemic graph. This ensures subsequent strategy cycles and related skills (vision, identity) can reference the decisions made.

node scripts/tortoise-memory.mjs write-points \
  --kind strategy \
  --points-json '[
    {"content": "<key decision 1>", "authoredBy": "define-strategy-skill", "confidence": 0.8},
    {"content": "<key constraint 2>", "authoredBy": "define-strategy-skill", "confidence": 0.9},
    {"content": "<open question>", "authoredBy": "define-strategy-skill", "confidence": 0.3}
  ]'

What to write (minimum):

• Each initiative as a strategy Point (confidence based on analysis depth)
• Key constraints and assumptions (high confidence)
• Explicit trade-offs ("we chose X over Y because Z")
• Resources allocated and why

Graceful degradation: If status is not_configured or tortoise_unavailable → skip with note.

What fails if you skip

If you skip...	This breaks...
Gap analysis	Strategy built on assumptions, not data. Competitor threats invisible.
Horizon mapping	Everything is urgent. No sequencing. Burnout and thrash.
Initiative generation	Vision never becomes action. Strategy is a document, not a plan.
Role gap analysis	Initiatives have no owners. Work falls through cracks.
Workflows plan	Roles overloaded or under-specified. Coordination overhead explodes.
Coherence review	Contradictions between steps. Strategy doesn't hang together.
Roadmap	Coordinator has nothing to execute. Strategy stays on the shelf.

---

Continue following the workflow as mandated by this skill. Do not skip steps.