@alphazede/bearing-lite 0.1.0

package/skills/bearing-lite/references/task-state.md

@@ -0,0 +1,42 @@
+# Task state (authoritative text)
+
+The project's human-readable plan is the only task-state record. Diagrams explain; they never authorize a transition, create state, or grant authority.
+
+## States and active owners
+
+| State | Active owner | Meaning |
+| --- | --- | --- |
+| `PROPOSED` | Parent coordinator | Task exists but is not execution-ready |
+| `READY` | Parent coordinator | Dependencies, scope, authority, and route are satisfied |
+| `WAITING_ON` | Parent coordinator | Named prerequisite or independent-assurance dispatch unavailable |
+| `IN_PROGRESS` | Assigned worker or coordinator | Assigned action is being performed |
+| `EVIDENCE_READY` | Parent coordinator | Candidate and evidence ready for next missing assurance |
+| `VALIDATING` | Validator | Evidence sufficiency under validation |
+| `REVIEWING` | Park Ranger when required | Independent defect review active |
+| `ACCEPTANCE` | Surveyor, Owner Authority, or parent coordinator when `required_assurance` is `none` | User-facing acceptance or coordinator completion confirmation is active |
+| `CORRECTION_REQUIRED` | Navigator or nearest parent coordinator | Agent-owned in-authority correction required |
+| `OWNER_DECISION_REQUIRED` | Owner Authority | Authority, scope, security, or replacement-path judgment required |
+| `COMPLETE` | Parent coordinator after assurance | Outcome and required assurance satisfied |
+| `CANCELLED` | Owner Authority or authorized parent | Task will not proceed |
+
+## Legal transitions
+
+- `PROPOSED` → `READY` when dependencies, scope, authority, assigned role, and assurance are declared and satisfied; else `WAITING_ON`.
+- `WAITING_ON` → `READY` when the prerequisite is repaired or replaced; → `EVIDENCE_READY` when fresh assurance becomes available; → `CANCELLED` on owner scope reduction.
+- `READY` → `IN_PROGRESS` when the assigned role starts; → `OWNER_DECISION_REQUIRED` on authority or integrity guard.
+- `IN_PROGRESS` → `EVIDENCE_READY` with candidate and handoff; → `CORRECTION_REQUIRED` on correctable failure; → `OWNER_DECISION_REQUIRED` on security, authority, or integrity guard.
+- `CORRECTION_REQUIRED` → `READY` on attempt 1 or 2 with a new hypothesis and evidence; → `OWNER_DECISION_REQUIRED` on the third failed correction or out-of-contract amendment.
+- `EVIDENCE_READY` → `VALIDATING` | `REVIEWING` | `ACCEPTANCE` for the next missing required role; → `WAITING_ON` if assurance dispatch is unavailable.
+- After accepted Validator or Park Ranger handoff, return to `EVIDENCE_READY` and select the next missing assurance role in declared order.
+- `VALIDATING` → `CORRECTION_REQUIRED` when more evidence or repair is needed; otherwise back through `EVIDENCE_READY` for remaining assurance.
+- `REVIEWING` → `CORRECTION_REQUIRED` | `EVIDENCE_READY` | `OWNER_DECISION_REQUIRED` by verdict.
+- `ACCEPTANCE` → `COMPLETE` when every required assurance accepted the same candidate; when `required_assurance` is `none`, parent-coordinator confirmation satisfies the assurance requirement; → `CORRECTION_REQUIRED` on acceptance gap.
+- `OWNER_DECISION_REQUIRED` → `READY` or `CANCELLED` by owner choice.
+
+## Ownership rules
+
+- One parent coordinator writes each task block and its transitions.
+- Navigator alone writes cross-wave dependencies and Expedition-wide sequencing.
+- Workers and assurance roles return handoffs; they do not race plan edits.
+- Candidate authors never provide their own Validator, Park Ranger, or Surveyor verdict.
+- Waiting on a prerequisite consumes no correction attempt. Each task has its own three-attempt correction counter; identical retries without new evidence are invalid.

package/skills/bearing-lite/references/task-state.mmd

@@ -0,0 +1,29 @@
+stateDiagram-v2
+    [*] --> PROPOSED
+    PROPOSED --> READY
+    PROPOSED --> WAITING_ON
+    WAITING_ON --> READY: prerequisite satisfied
+    WAITING_ON --> EVIDENCE_READY: fresh assurance available
+    WAITING_ON --> CANCELLED
+    READY --> IN_PROGRESS
+    READY --> OWNER_DECISION_REQUIRED
+    IN_PROGRESS --> EVIDENCE_READY
+    IN_PROGRESS --> CORRECTION_REQUIRED
+    IN_PROGRESS --> OWNER_DECISION_REQUIRED
+    CORRECTION_REQUIRED --> READY
+    CORRECTION_REQUIRED --> OWNER_DECISION_REQUIRED: third failed correction
+    EVIDENCE_READY --> VALIDATING: Validator required
+    EVIDENCE_READY --> REVIEWING: Park Ranger required
+    EVIDENCE_READY --> ACCEPTANCE: Surveyor, owner, or coordinator when required_assurance is none
+    EVIDENCE_READY --> WAITING_ON: assurance dispatch unavailable
+    VALIDATING --> CORRECTION_REQUIRED
+    VALIDATING --> EVIDENCE_READY: pass, select next assurance
+    REVIEWING --> CORRECTION_REQUIRED
+    REVIEWING --> EVIDENCE_READY: accepted, select next assurance
+    REVIEWING --> OWNER_DECISION_REQUIRED
+    ACCEPTANCE --> CORRECTION_REQUIRED
+    ACCEPTANCE --> COMPLETE
+    OWNER_DECISION_REQUIRED --> READY
+    OWNER_DECISION_REQUIRED --> CANCELLED
+    COMPLETE --> [*]
+    CANCELLED --> [*]

package/skills/bearing-lite/templates/task.md

@@ -0,0 +1,55 @@
+# Task record template
+
+Use one compact Markdown block per task in the project's existing plan or progress document. Omit unused conditional fields. Prefer the project's format; add only missing fields. Do not create a Bearing-owned state file or hidden ledger.
+
+## Always present
+
+```markdown
+### task_id: T1
+- outcome: <approved user-visible outcome for this task>
+- status: PROPOSED
+- assigned_role: <role or unassigned>
+- depends_on: []
+- next_action: <smallest concrete next step>
+```
+
+Field rules:
+
+- `task_id` — stable identifier unique within the plan.
+- `outcome` — approved intent for this task only.
+- `status` — one of `PROPOSED`, `READY`, `WAITING_ON`, `IN_PROGRESS`, `EVIDENCE_READY`, `VALIDATING`, `REVIEWING`, `ACCEPTANCE`, `CORRECTION_REQUIRED`, `OWNER_DECISION_REQUIRED`, `COMPLETE`, `CANCELLED`.
+- `assigned_role` — current role contract for the active step, or unassigned while proposing.
+- `depends_on` — list of other `task_id` values only, never prose.
+- `next_action` — the immediate allowed step.
+
+## Before execution (add when leaving PROPOSED)
+
+```markdown
+- scope: <allowed paths, systems, and boundaries>
+- authority: <approved envelope; protected actions remain owner-only>
+- required_assurance: [Validator]
+```
+
+`required_assurance` lists only roles that must accept the same candidate (for example `Validator`, and `Park Ranger` or `Surveyor` when their triggers apply).
+
+## After candidate work (add when evidence exists)
+
+```markdown
+- candidate_ref: <strongest native revision or changed-path reference available>
+- evidence: <commands run, results observed, and inferences labeled separately>
+```
+
+`candidate_ref` must not claim stronger provenance than the client can prove.
+
+## Waiting or correcting only
+
+```markdown
+- blocker: <WAITING_ON:<task-id> | typed reason | none on success>
+- attempts: <0-3 correction attempts for this task or protected action>
+```
+
+Omit `blocker` and `attempts` when the task is not waiting or correcting. Waiting does not consume correction attempts.
+
+## Single-writer reminder
+
+Only the parent coordinator updates this block after rereading it. Crewmate, Validator, Park Ranger, and Surveyor return handoffs; the coordinator records transitions. Navigator alone changes cross-wave dependencies or global sequencing.

package/skills/crewmate/SKILL.md

@@ -0,0 +1,37 @@
+---
+name: crewmate
+description: >
+  Implement one approved mutation, execution, or integration packet inside its
+  exact write set with native tools. Use for crewmate, implement packet, bounded
+  coding task, or apply a write-set change. Do not use for wave orchestration,
+  multi-packet control, self-certification, publication, or expanding authority.
+---
+
+# Crewmate
+
+## Trigger
+
+- **Match:** one approved implementation, repair, or integration packet is READY with fixed scope and authority.
+- **Non-match:** multi-packet orchestration, wave ownership, assurance review, owner decisions, or unset write sets.
+
+## Inputs
+
+Plan path, packet objective, acceptance, allowed paths, commands, stop condition, `required_assurance`, expected handoff.
+
+## Responsibility
+
+Smallest honest change in the write set. For verified contract bugs, add a failing regression first, then repair. Run assigned focused commands. Preserve unrelated work.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `CANDIDATE_READY`, `PARTIAL`, `WAITING_ON`, `OWNER_DECISION_REQUIRED`. Receiver: parent coordinator; Validator only when `required_assurance` includes it or owner opts in.
+
+## Independence
+
+Never self-certifies. `required_assurance: none` means author self-check plus coordinator confirmation—not automatic Validator/Park Ranger.
+
+## Correction
+
+Repair evidenced failures inside authority. Design, write-set, or acceptance changes escalate without silent expansion.

package/skills/delegate-authority/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: delegate-authority
+description: >
+  Maintain owner-approved multi-phase intent and appoint phase Navigators when
+  the owner explicitly delegates across sessions or Navigator replacement. Use
+  for delegate authority, multi-phase journey, or long-running delegation. Do
+  not use for single-wave work, implementation, assurance, contract changes, or
+  publication.
+---
+
+# Delegate Authority
+
+## Trigger
+
+- **Match:** owner explicitly delegates a bounded multi-phase journey, especially across sessions or Navigator replacement.
+- **Non-match:** ordinary direct, Explorer, or single-phase Expedition work without owner delegation.
+
+## Inputs
+
+Plan path, approved intent, phase boundaries, authority envelope, `required_assurance`, expected handoff.
+
+## Responsibility
+
+Preserve approved intent, appoint phase Navigators, and route phase handoffs. Never implement, assure, amend the contract, or publish.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `READY`, `REROUTED`, `WAITING_ON`, `OWNER_DECISION_REQUIRED`. Receiver: Navigator or Owner Authority.
+
+## Independence
+
+Dormant until owner delegation is evidenced. Does not self-certify phases.
+
+## Correction
+
+In-authority phase appointment repair only. Contract, security, or publication changes escalate to Owner Authority.

package/skills/explorer/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: explorer
+description: >
+  Wave-owner orchestration: sequence and dispatch bounded packets inside one
+  approved wave without implementing them. Use for explorer, wave ownership,
+  packet orchestration, or parallel dispatch planning. Do not use for product
+  implementation, multi-wave Trail Boss control, nested Sub-explorer defaulting,
+  or independent assurance of wave output.
+---
+
+# Explorer
+
+## Trigger
+
+- **Match:** an approved wave contains work packets that need orchestration, sequencing, or parallel dispatch (not the implementation itself).
+- **Non-match:** a single already-bounded implementation packet with no wave orchestration need; multi-wave Trail Boss duty; assurance review.
+
+## Inputs
+
+Plan path, wave objective, packets, dependencies, scope, authority, `required_assurance`, expected handoff.
+
+## Responsibility
+
+Issue disjoint Crewmate packets, inspect returns against allowlists, and hand off for coordinator confirmation or required assurance. Never implement or audit the packets itself.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `READY`, `REROUTED`, `WAITING_ON`, `OWNER_DECISION_REQUIRED`. Receiver: Sub-explorer, Crewmate, or parent coordinator.
+
+## Independence
+
+Wave owner only. Does not provide Validator, Park Ranger, or Surveyor verdicts for candidates produced under its dispatch when it is author-equivalent.
+
+## Correction
+
+Reroute incomplete packets with new evidence. Authority or design amendments escalate to parent coordinator or Owner Authority.

package/skills/gather-supplies/SKILL.md

@@ -0,0 +1,33 @@
+---
+name: gather-supplies
+description: >
+  Resolve only material owner decisions that block a complete plan after the
+  workspace exists. Use for gather supplies, owner decisions, plan-spec
+  assumptions, or blocking scope questions. Do not use for design drafting,
+  implementation, product edits, or replaying settled owner answers.
+---
+
+# Gather Supplies
+
+Procedural stage. Not a persona.
+
+## Match / non-match
+
+- **Match:** material open decisions block scope, architecture, security, authority, or acceptance.
+- **Non-match:** decisions already recorded; Map the Route can proceed; pure implementation packets.
+
+## Inputs
+
+Goal, repository evidence, plan workspace, and prior owner Q&A.
+
+## Procedure
+
+1. Ask only material questions; reuse durable prior answers without re-asking.
+2. For each question: recommendation, evidence, options/tradeoffs, affected section, safe default.
+3. Apply completed answers into the validated plan specification only.
+4. Keep requirements testable with stable IDs; record blockers instead of inventing approval.
+5. Exit with zero-to-few bounded questions, applied plan-spec, or `OWNER_DECISION_REQUIRED`.
+
+## Never
+
+Design the route, draft implementation slices, edit product code, or overwrite a recorded owner answer with a default.

package/skills/map-the-route/SKILL.md

@@ -0,0 +1,33 @@
+---
+name: map-the-route
+description: >
+  Build the human-readable task graph when the outcome is approved but
+  executable tasks are missing. Use for map the route, implementation slices,
+  task graph, SEIT, or planning artifacts. Do not use to execute work, select
+  models, self-certify assurance, or advance journey state beyond planning.
+---
+
+# Map the Route
+
+Procedural stage. Not a persona.
+
+## Match / non-match
+
+- **Match:** plan-spec/decisions are approved and executable task graph or planning package is incomplete.
+- **Non-match:** tasks already READY with roles and assurance; pure Crewmate packet work.
+
+## Inputs
+
+Approved goal, decisions, plan-spec, and required Role routes.
+
+## Procedure
+
+1. Author only missing planning artifacts: design, SEIT, and implementation slices as needed.
+2. For each task declare `depends_on`, assigned role, scope, authority, and `required_assurance` (explicit `none` allowed).
+3. Prefer the smallest valid route; leave dormant roles unselected.
+4. Keep IDs stable; cite acceptance and evidence expectations per task.
+5. Return artifact paths, blockers, and next planning or execution handoff without executing.
+
+## Never
+
+Implement product changes, invent owner approval, run phase gates as automatic, or claim independent assurance.

package/skills/map-the-route/references/artifact-grammar.md

@@ -0,0 +1,83 @@
+# Map the Route Artifact Grammar
+
+Apply this closed grammar while authoring or validating planning artifacts.
+Bearing rejects deviations with typed findings.
+
+## Shared artifact rules
+
+1. Frontmatter declares `type` as `plan-spec`, `design`, `seit`, or
+   `implementation`, and `status` as `complete` or `amended`.
+2. Requirement IDs use `AC-*` or `RISK-*`; design IDs use `DES-*` or
+   `CONTRACT-*`; SEIT rows use `SEIT-*`; commands use `CMD-*` or `PROC-*`.
+   Suffixes contain only uppercase letters, digits, dots, or hyphens.
+3. `plan-spec.md` declares requirements, Entry criteria, Exit criteria,
+   Rollback or repair, and Accountable controller. `design.md` declares design
+   IDs. `seit.md` declares SEIT rows and commands.
+4. Preserve stable IDs. Every implementation reference resolves to a declared
+   requirement, design contract, SEIT row, and command.
+5. The plan's Role routes decision supplies immutable `authorRoute` and
+   `reviewSlots`. Missing or incomplete routes block implementation drafting.
+
+## SEIT rules
+
+1. Include non-empty `Required Commands`, `Traceability Matrix`, and
+   `Cross-cutting Checks` sections.
+2. Declare each command as `- **<CMD/PROC-id>** — description`.
+3. Use one flat traceability table with exactly these columns: SEIT row ID |
+   Acceptance/risk ID | Design/contract ID | Boundary/test layer | Positive
+   case | Negative/failure case | Command/procedure ID | Evidence.
+4. Every row carries exactly one SEIT row ID, requirement ID, design ID, and
+   command ID, and names an observable failure.
+5. If procedure narratives are used, title each `### SEIT-<id> <title>` under
+   a procedures section and restate Command, Positive case, Negative case, and
+   Evidence exactly. Every PROC row then has exactly one matching narrative.
+
+## Implementation rules
+
+1. Slice headings are `### Slice <id>` and manifests are
+   `### <id> execution manifest`. IDs are a letter-run plus an integer or
+   dotted integer. Every slice has exactly one matching manifest.
+2. Every slice declares Goal, Requirement IDs, Design IDs, SEIT proof rows,
+   Type, Design lenses, Implementation role, Agent model route, Agent reasoning
+   level, and Review path. Goals are at most 512 characters.
+3. Every manifest declares Write set, Command IDs, Stop condition, and Human
+   decision. Optional fields are Shared interfaces (`path#Symbol`), Integration
+   boundary, and Parallel safe (`yes` or `no` plus reason).
+4. Write sets use one line: `Write only `path``. Paths are bounded, normalized,
+   repository-relative literals. Put prohibitions in prose, not the write set.
+5. Multi-slice plans declare consecutive `Wave <n>: <ids>` lines. Every slice
+   belongs to one wave. Dependencies use acyclic `S1 --> S2` arrows.
+6. Optional phase graphs use Phase | Slices | Depends on phases | Integration
+   checkpoints. Optional Inputs and Produces name metric denominator, ledger
+   key, or contract field values in backticks.
+7. Plans may contain at most 128 slices, manifests, write paths, and commands.
+   Aim for at most 500 estimated tokens per slice plus manifest; split larger
+   packets when practical.
+
+## Optional binding sections
+
+If `## System Catalog` appears, use System ID | System | Responsibility with
+one `SYS-*` per row. Each `### SYS-*` specification declares Ownership, Inputs,
+Outputs, APIs, Data ownership, Invariants, Trust boundary, Failure modes, and
+Observability. A Requirement Trace, when present, maps Requirement ID, System
+ID, Contract ID, SEIT row ID, Slice ID, and Path; every requirement and path
+must resolve.
+
+If `## Risk Profile` appears, enumerate the complete closed flag set required
+by Bearing. Every `yes` maps design or system, SEIT row, and slice coverage.
+Every `no` gives an evidence-backed rationale of at least four words; do not use
+placeholders, bare negations, or deferral language.
+
+## Feature diagram and visual review rules
+
+1. Select review-oriented feature diagrams (flow diagrams, state-machine diagrams, or process/sequence diagrams) when they materially clarify architecture or lifecycle behavior for owner review.
+2. Major features select a justified subset or full set of diagrams based on architectural complexity. Trivial features or minor fixes do not require unnecessary diagrams.
+3. Every feature diagram must include canonical, reviewable source (such as Mermaid code blocks), render visually inside generated `review.html`, and retain nearby authoritative explanatory text.
+
+## Completion boundary
+
+Design-only work returns after valid `design.md` and `seit.md`. Implementation
+drafting returns after valid `implementation.md`. Complete-route work remains
+incomplete until `plan-spec.md`, `design.md`, `seit.md`, and `implementation.md`
+are current. Bearing then generates the complete `review.html`; agents never
+hand-edit, summarize, or replace it.

package/skills/navigator/SKILL.md

@@ -0,0 +1,44 @@
+---
+name: navigator
+description: >
+  Expedition and multi-phase Navigator sequencing across dependency-connected
+  waves, including gap correction and in-authority coverage expansion. Use for
+  navigator, expedition integration boundary, or multi-phase wave graph control.
+  Do not use for single bounded packets, product implementation under Navigator
+  identity, or independent assurance of own corrections.
+---
+
+# Navigator
+
+## Trigger
+
+- **Match:** work spans multiple phases, dependency-connected waves, or an Expedition integration boundary.
+- **Non-match:** one packet or one Explorer-owned wave with no cross-wave sequencing need.
+
+## Inputs
+
+Plan path, wave graph, dependencies, approved outcome, authority, allowed paths, `required_assurance`, expected handoff.
+
+## Responsibility
+
+Sequence Explorers (and Trail Boss only when multi-wave conflict requires it), correct in-authority gaps, and integrate returns. Use one fresh Navigator context per execution wave; after phase acceptance PASS, return to Delegate Authority rather than continuing into the next wave. A fresh wave Navigator rereads `implementation.md`, `review.html`, and `seit.md` before acting. Never implement under Navigator identity, supply independent assurance, publish, release, mutate remotes, or take destructive action under Navigator identity.
+
+## Authority
+
+Approved plan plus task authority is sufficient for execution, routing, bounded repairs, and in-contract coverage amendments. Do not re-request Owner Authority for work already authorized by that plan.
+
+Before any `OWNER_DECISION_REQUIRED` return, seek and attempt a safer in-authority workaround when one exists. Ask Owner Authority only when outcome or acceptance meaning changes; repository or product boundary changes; public, publication, remote, destructive, release, or release-target action is needed; secret handling or a security exception changes; an external dependency or cost requires new authority; a confirmed owner decision would change; any other authority envelope would expand; or the declared third failed correction occurs.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `READY`, `REROUTED`, `WAITING_ON`, `OWNER_DECISION_REQUIRED`. Receiver: Trail Boss, Explorer, assurance role when required, Delegate Authority, or Owner Authority.
+
+## Independence
+
+Writes only coordinator sequencing and task blocks it owns. Does not supply Validator, Park Ranger, or Surveyor verdicts for candidates it integrated as author-equivalent.
+
+## Correction
+
+Attempts 1–2 need new hypothesis and evidence. Third failure or out-of-authority change → `OWNER_DECISION_REQUIRED`.

package/skills/park-ranger/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: park-ranger
+description: >
+  Review introduced defects on one exact candidate when required_assurance
+  includes Park Ranger, the owner opts in, or a material integrated phase gate
+  requires defect review. Use for park ranger, defect review, or introduced-bug
+  adjudication. Do not use for implementation, automatic per-slice gates,
+  sufficiency scoring, or user-facing acceptance.
+---
+
+# Park Ranger
+
+## Trigger
+
+- **Match:** `required_assurance` lists Park Ranger; owner opts in; or integrated phase gate requires defect review on a stable candidate.
+- **Non-match:** `required_assurance: none` without owner opt-in/phase gate; Validator sufficiency-only; Surveyor acceptance; implementation packets.
+
+## Inputs
+
+Plan path, exact `candidate_ref`, diff or evidence, prior Validator result when present, expected handoff.
+
+## Responsibility
+
+Find introduced defects with reproduction and reachability. Rank findings. Never implement fixes or repeat an unchanged review as new proof.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `BLOCK`, `REPAIR_REQUIRED`, `ACCEPT_WITH_FINDINGS`, `ACCEPT`. Receiver: parent coordinator, Surveyor when next required, or correction owner.
+
+## Independence
+
+Fresh non-author only. Never automatic per slice. Candidate authors never supply this verdict.
+
+## Correction
+
+Returns repair targets only. Does not edit the candidate or re-review without new evidence.

package/skills/repository-fit/SKILL.md

@@ -0,0 +1,33 @@
+---
+name: repository-fit
+description: >
+  Propose one target repository and plan-directory assumption when the target
+  is absent or ambiguous. Use for repository fit, choose repo, or confirm
+  workspace before Set Bearings. Do not use after fit is confirmed, for product
+  edits, routing roles, or owner-only publication decisions.
+---
+
+# Repository Fit
+
+Procedural stage. Not a persona.
+
+## Match / non-match
+
+- **Match:** selected repository is missing, multi-root, or identity is ambiguous.
+- **Non-match:** owner already confirmed fit; Set Bearings or later stages are ready.
+
+## Inputs
+
+Work goal, candidate roots, and any prior fit evidence.
+
+## Procedure
+
+1. Inspect only the selected root; open one extra root only if owner-authorized.
+2. Cap discovery (depth 4, 200 paths). Prefer manifests, top-level docs, plan conventions.
+3. Propose exactly one repository and one plan-directory assumption with path citations.
+4. Ask one concise owner confirmation question. Do not treat a recommendation as approval.
+5. Return `assumption` + `question`, or typed stop: `fit_unavailable`, `fit_malformed`, `fit_undecidable`.
+
+## Never
+
+Write files, walk parents without authority, rank alternatives after undecidable, or advance into Set Bearings.

package/skills/set-bearings/SKILL.md

@@ -0,0 +1,33 @@
+---
+name: set-bearings
+description: >
+  Create or resume the bounded plan workspace and repository map when project
+  context or current-state map is missing after fit. Use for set bearings,
+  plan workspace, or repository map. Do not use before fit is confirmed, for
+  owner decisions, design drafts, product edits, or risk-profile authoring.
+---
+
+# Set Bearings
+
+Procedural stage. Not a persona.
+
+## Match / non-match
+
+- **Match:** fit is confirmed and plan workspace or repository map is missing or stale.
+- **Non-match:** workspace already ready; Gather Supplies or later stages apply.
+
+## Inputs
+
+Repository root, work goal, and owner-confirmed plan directory (verbatim).
+
+## Procedure
+
+1. Use the confirmed plan directory exactly; never derive, slug, or suffix it.
+2. Create or resume the workspace without deleting existing plan content.
+3. Write only the plan-directory stub and sibling repository map.
+4. Leave risk-profile content to Gather Supplies / owner; do not invent it.
+5. Return relative paths and whether the workspace was created or resumed.
+
+## Never
+
+Ask planning questions, design, implement, edit product code, or delete plans or hidden runtime state.

package/skills/sub-explorer/SKILL.md

@@ -0,0 +1,37 @@
+---
+name: sub-explorer
+description: >
+  Orchestrate bounded packets inside one Explorer-owned wave when at least two
+  proven-independent lanes need one extra coordination level. Use for
+  sub-explorer or in-wave lane coordination. Do not use as a second nesting
+  level, for product implementation, multi-wave control, or assurance.
+---
+
+# Sub-explorer
+
+## Trigger
+
+- **Match:** one Explorer-owned wave has ≥2 proven-independent lanes and one extra coordination level materially reduces context or conflict.
+- **Non-match:** single lane; nested Sub-explorer requests; Trail Boss multi-wave work; direct Crewmate packets.
+
+## Inputs
+
+Parent wave plan path, lane boundaries, packet specs, dependencies, authority, `required_assurance`, expected handoff.
+
+## Responsibility
+
+Orchestrate in-wave packets only. Maximum one Sub-explorer level. Never perform packets, expand parent wave authority, or assure results.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `READY`, `REROUTED`, `WAITING_ON`, `OWNER_DECISION_REQUIRED`. Receiver: Crewmate or Explorer.
+
+## Independence
+
+Dormant unless independence is proven. Does not substitute for Explorer, Trail Boss, or Navigator.
+
+## Correction
+
+Return lane conflicts to Explorer. Out-of-wave expansion escalates to parent coordinator.

package/skills/surveyor/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: surveyor
+description: >
+  Compare user-facing behavior of a stable integrated candidate with the
+  approved outcome at multi-packet phase, Expedition, Final Audit, or
+  owner-selected acceptance boundaries. Use for surveyor, acceptance review, or
+  outcome comparison. Do not use for implementation, per-packet sufficiency,
+  introduced-defect triage as Park Ranger, or owner-only publication approval.
+---
+
+# Surveyor
+
+## Trigger
+
+- **Match:** multi-packet phase, Expedition, Final Audit, or owner-selected acceptance boundary has a stable integrated candidate.
+- **Non-match:** single unfinished packet; Validator-only sufficiency; Park Ranger defect pass; owner publication decisions reserved to Owner Authority.
+
+## Inputs
+
+Plan path, exact integrated `candidate_ref`, approved outcome, user-facing evidence, prior required assurance results, expected handoff.
+
+## Responsibility
+
+Read-only acceptance comparison against the approved outcome. Report gaps with precise locations. Never repair implementation or approve owner-only actions.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `ACCEPT`, `GAPS`, `OWNER_DECISION_REQUIRED`. Receiver: Owner Authority or completion transition via parent coordinator.
+
+## Independence
+
+Fresh non-author only. Distinct from Validator sufficiency and Park Ranger defect review. Authors never self-accept.
+
+## Correction
+
+Returns gaps only. Does not implement repairs or convert advice into owner publication authority.

package/skills/trail-boss/SKILL.md

@@ -0,0 +1,38 @@
+---
+name: trail-boss
+description: >
+  Trail Boss cross-wave coordination for concurrent Explorer-owned waves with
+  shared-interface or dependency conflicts. Use for trail boss, multi-wave
+  conflict management, wave readiness scheduling, or integration order across
+  waves. Do not use for a single wave, packet implementation, review, grading,
+  or assurance verdicts.
+---
+
+# Trail Boss
+
+## Trigger
+
+- **Match:** ≥2 Explorer-owned waves are active or need cross-wave dependency, shared-interface, or conflict management.
+- **Non-match:** one Explorer owns the whole route; direct Crewmate work; review or implementation tasks.
+
+## Inputs
+
+Plan path, wave graph, lane status, budgets, shared interfaces, `required_assurance`, expected handoff.
+
+## Responsibility
+
+Schedule wave readiness and integration order only. Never perform a wave's work, implement product changes, or fill review/assurance slots.
+
+## Return
+
+Handoff fields: `plan_ref`, `role`, `subject`, `depends_on`, `scope`, `authority`, `outcome`, `evidence`, `blocker`, `next_action`, `receiving_role`.
+
+Outcomes: `READY`, `REROUTED`, `WAITING_ON`, `OWNER_DECISION_REQUIRED`. Receiver: Explorer, Navigator, or assurance role when required.
+
+## Independence
+
+Orchestration-only. Omitted when one Explorer suffices. Never self-certifies integrated candidates.
+
+## Correction
+
+Escalate unresolvable cross-wave conflicts to Navigator or Owner Authority; do not absorb Explorer packet work.