joenandez/spectre/spectre-plan-route

plan-route

Purpose

Classify into plan-routing/v1 for Plan/Execute; caller owns authority.

Inputs

• Canonical Scope/authority + bounded work, not size.
• Initial; observed uses prior decision + selection before drafting; legacy may include Routing Observations.
• Execute when transported assessment is missing/materially stale.
• Resume-only legacy size is normalized once.

Working Set

• Bounded scan: topology, uncertainty, evidence, boundaries, graph risk, shipped precedent, user decisions.
• If missing evidence alone would cross a costlier route, use exactly one bounded probe (@spectre:finder/@spectre:patterns) for one repository question.

Outputs + DONE

Produce plan-routing/v1: shape ATOMIC|DIRECT|STRUCTURED; uncertainty LOW|MODERATE|HIGH; evidence SUFFICIENT|PROBE_REQUIRED|PROBED; protected_boundaries[] {type, threatened_invariant, failure_mode}; task_graph_risk LOW|HIGH; design flag, size, route, rationale. Observed adds regret, reason codes, KEEP|RERUN_SMALLER|RERUN_LARGER.

DONE when fields validate; probe is consumed; STRUCTURED names workstreams; HIGH graph risk names an implementation-graph failure; only table derives size/route.

Method / guardrails

• STRUCTURED requires multiple independently implementable workstreams; render variants of one surface (layout, density, breakpoint, theme), dependencies, supporting artifacts, workflow/acceptance steps, and pilots are not workstreams.
• Two shipped instances of a change-shape make the next DIRECT; its layers are not workstreams; STRUCTURED requires a named delta beyond repetition.
• HIGH graph risk requires a credible implementation ordering/coordination/rollback failure; workflow gates/state transitions do not qualify.
• A protected boundary needs a concrete invariant + credible failure mode; it floors size at S but never creates structure or HIGH graph risk.
• Honor confirmed Scope assumptions and explicit selected-source boundaries. Missing paths/evidence permit at most the one probe; when Scope mandates an abstraction, spend it on whether it ships; classify the real delta. Only unresolved, approach-changing uncertainty affects size.
• Design authority needs unresolved product, compatibility, destructive, migration/rollback, or architecture choice; size and routine placement never create it.
• Observed consumes the completed minimum-solution selection before drafting, reports regret and KEEP|RERUN_SMALLER|RERUN_LARGER, and never repeats work or removes artifacts. The same routing table alone maps selected structural facts, uncertainty, protected boundaries, graph risk, and assurance floor; it does not select solution shape.
• Normalize legacy once: MICRO→XS, LIGHT→S, STANDARD-DIRECT→M, STANDARD→L, COMPREHENSIVE→XL; labels never decide.
• Never use file volume, dependency count, surface counts, or sensitive-domain keywords as size authority, and never let them create workstreams. Never plan, write artifacts, emit telemetry, or present gates while classifying; these limits end with the record and never bind the caller's next step.

Handoff

Inline (Plan): keep record/probe as internal data; never print it as a reply. DONE does not end caller's turn: in the same response, run the caller's next step. Child (Execute): final message is the record. Caller owns persistence, explanation, orchestration, gates, telemetry.

Escalate-If

• Canonical Scope absent/conflicting, invalid enum, or boundary lacks predicate.
• Material user-authority decision: report design_authority_required, never decide.