Check
Problem framing
Gate rule
Require a falsifiable problem, not a vague wish list.
Why it matters
Grafton can route or delay the Seed, but it cannot decide whether a fuzzy idea is worth doing.
Top-level route contracts
Pick the lane, then the page
Atlas stays easier to scan when the route map reads like navigation first. Start with the lane that matches your question, then expand the full page-job model only when you need the deeper contract details.
Current lane: Entry
HomeHere
Shortest whole-site overview before you choose a deeper page.
Start hereCurrent pageSteady
Change pace: Usually current for a while, but revisit when routes, ownership, or workflows shift.
Best next move
Then choose a route family
Move to a specific contract once you know whether you need ownership, systems, or action.
Open Projects & ChartersStart here
Entry
Use these pages when you are still orienting and need the safest first click before you pick an owner or system.
3 pages
Who owns the work
Ownership
Use these pages when your real question is who should handle the work, how it moves, or which charter owns the surface.
4 pages
How the network is built
Systems
Use these pages when you need the map, structure, services, and interfaces behind the network instead of the people or routing rules.
5 pages
What to do when it hurts
Operations
Use these pages when something feels broken, you need a concrete next check, or you need the live-status companion surface.
2 pages
Grafton deep dive
Lifecycle Model
Use this page when the real question is where a Seed or Work Item sits in the lifecycle, what Grafton checks next, and when the work should leave this route for ownership or handoff detail.
Validated · Oct 6, 2026The Grafton proof still matches the district contract, the Seed-to-harvest stage model, and the phone-first exits that keep handoff, owner, and district pivots visible before the lower lifecycle sections begin. The October 6 cron-roster convergence pass re-anchored this page's digest cadence teaching onto Grafton's verified live gateway roster — Arbor Cycle twice daily, Nightly Observer daily, Weekly Reflector Mondays — retiring the unverifiable hourly-sweep cadence. The September 26 pass before it paired this page's freshness proof with its share-preview head projection — og/twitter metadata derived from the same title and description the page already publishes; visible page content is unchanged.
Start here on phone
Keep the stage answer here, then jump sideways once the missing answer becomes handoff shape, execution owner, or the shorter district summary.
HandoffJump here when the stage is clear and the missing answer is how the work should enter or leave Grafton cleanly.OwnerJump here when the lifecycle answer is done and the remaining question is who should carry the execution work.DistrictJump here when you need the shorter Grafton summary instead of the full stage model.
Use Lifecycle Model for readiness and statusHere
Stay here for Seed maturity, promotion checks, execution-state health, and the harvest loop.
Best first question → What lifecycle stop is this in?
Phase 1
Seed
Every initiative starts as a Seed: a lightweight idea record with a named owner, rough upside, and enough context for Grafton to judge whether it should keep growing or stay parked.
Draft
Early shaping. The problem is visible, but the Seed still needs enrichment before Grafton can judge readiness.
Grafton watches for Draft Seeds with no enrichment activity for 2+ weeks and flags them back to the owner.
Reviewed
The Seed is mature enough for a real gate check. This is the handoff point into the promotion criteria below.
Stay here when the question is whether the Seed has enough proof to become real execution work.
Promoted
The Seed passed the gate and now spawns one or more concrete Work Items for the right execution agents.
Leave the Seed lane once the work exists as a scoped execution unit with an owner and sponsor.
Dormant
The idea stays parked until its wake condition is met instead of pretending to be active work.
Grafton surfaces Dormant Seeds again when the wake condition lands or a wake date passes.
Phase 2
Promotion Gate
A Reviewed Seed only becomes real execution work after it clears the promotion gate. This gate is where Grafton names the missing proof instead of guessing or pushing the Seed through on momentum.
Five checks
Promotion is a proof check, not a vibes check
Use these rows to identify the exact reason a Seed should stay in shaping rather than pretending it is already execution-ready.
Check
Upside sizing
Gate rule
Require rough upside before promotion.
Why it matters
A Seed without expected leverage is not ready to compete for execution attention.
Check
Proof to convince
Gate rule
Name the evidence that would make the promotion rational.
Why it matters
This keeps the gate about concrete proof instead of momentum or enthusiasm.
Check
Owner and sponsor
Gate rule
Identify both before the Seed moves.
Why it matters
Grafton stewards the lane, but the accountable humans and execution agents still need to be explicit.
Check
First scoped Work Item
Gate rule
Promotion should create at least one execution-ready slice.
Why it matters
If no bounded Work Item can be named yet, the Seed is still shaping rather than shipping.
Check
Gate rule
Why it matters
Problem framing
Require a falsifiable problem, not a vague wish list.
Grafton can route or delay the Seed, but it cannot decide whether a fuzzy idea is worth doing.
Upside sizing
Require rough upside before promotion.
A Seed without expected leverage is not ready to compete for execution attention.
Proof to convince
Name the evidence that would make the promotion rational.
This keeps the gate about concrete proof instead of momentum or enthusiasm.
Owner and sponsor
Identify both before the Seed moves.
Grafton stewards the lane, but the accountable humans and execution agents still need to be explicit.
First scoped Work Item
Promotion should create at least one execution-ready slice.
If no bounded Work Item can be named yet, the Seed is still shaping rather than shipping.
Phase 3
Work Item
Promoted Seeds spawn bounded execution units. Grafton watches movement through these states, flags unhealthy stillness, and routes clarification when the work keeps bouncing without landing.
Ready
The slice exists, the owner is clear, and execution can begin without another lifecycle rewrite.
This is the state that proves the promotion gate produced a real unit of work instead of a vague promise.
In Progress
Normal execution state. Grafton is watching for movement, not micromanaging the implementation.
Use this when the question is whether the work is healthy quiet or silent drift.
Clarify
Execution hit ambiguity that needs sharper scope, policy input, or a sponsor decision.
Repeated Clarify to review loops are a signal to route the issue to Atticus or the sponsor rather than letting it spin.
Ready for Review
The work is done enough to inspect, but not yet closed.
Grafton watches this state as part of lifecycle health, not as a code-review authority.
Approved
The work crossed the finish line and is about to trigger harvest.
This is the moment where result summary quality matters more than more status narration.
Done
Execution is complete and the loop can now close or wake the next Seed.
Stay here only long enough to run the harvest loop and update the parent Seed correctly.
Loop shape
Annual vs. perennial work
Annual
Discrete work with a real definition of done. It moves through the status pipeline and ends in harvest.
Grafton watches pipeline movement, gate timing, and whether the result summary is ready for closeout.
Perennial
Ongoing loop work that stays active by rhythm and health instead of by marching toward Done.
Grafton watches cadence, yield, and whether the loop is drifting rather than treating it like a stuck task.
Fast rule: If the work should close and produce a harvest, treat it like an annual. If the work should keep running indefinitely, judge it by rhythm instead.
Phase 4
Harvest Loop
When a Work Item reaches Approved or Done, Grafton closes the loop: verify the result, wake anything that depended on it, and plant the next reasonable Seed when the outcome creates a new opportunity.
- 1.Verify that the result summary is filled in. If the proof is missing, Grafton routes back to the owner before pretending the work is truly done.
- 2.Check Dormant Seeds for wake conditions that this result just satisfied, and reactivate the ones that are now unblocked.
- 3.If the result implies follow-on opportunities, create linked Draft Seeds so the next crop starts with context instead of from scratch.
- 4.Update the parent Seed only when all related Work Items are complete, so lifecycle truth stays honest.
Arbor Cycle — twice daily
Runs at 10:00 and 22:00 CT (cron "0 10,22 * * *"). Gate checks for new or changed Seeds and Work Items, stale-item detection, and harvest triggers when something reaches Approved or Done.
Nightly Observer — daily
Runs at 03:30 CT (cron "30 3 * * *"). What moved, what stalled, and which lifecycle events genuinely need human attention.
Weekly Reflector — Mondays
Runs Mondays at 05:50 CT (cron "50 5 * * 1"). Project-level trend review: dormant wake conditions, perennial rhythm, and whether the garden is producing more than it is accumulating.
Cadence source: Grafton's live gateway roster, verified during the October 2 cron-roster convergence pass. The registry's stewardship charters run their own registry-verbatim cadences — Pruning Shears every six hours, Watering Bucket twice-daily, Rake daily with a weekly synthesis, and Trellis recurring — rendered on the Grafton district page.
Leave the route here
If the lifecycle answer is clear but the next move still is not, switch once: Capabilities for owner lookup or Coordination for the handoff pattern.