owlrunkit 0.15.1 → 0.16.1

package/integrations/codex/skills/owlcoda-runkit/SKILL.md

@@ -7,8 +7,15 @@ description: Run receipt-backed, project-owned software execution with OwlCoda R
 
 Use the bundled deterministic Core to coordinate work through project artifacts rather than chat memory. Require Node.js 20 or later.
 
-Current release: standalone `owlrunkit@0.15.1`, bundling Contract `0.2` and
-Core `0.15.1`. The root `owlcoda` package has a separate version lifecycle.
+Bundled release: standalone `owlrunkit@0.16.1`, bundling Contract `0.2` and
+Core `0.16.1` with manifest
+`sha256:38ac9110e328c38a81db05f1359734ca810b991c9f4de77647f9719f5e6af78b`.
+Publication and dist-tag state are mutable registry facts: verify the exact
+official npm registry provenance before adoption. The immutable rollback
+baseline for this release is `owlrunkit@0.16.0` with SHA-1
+`bab651b0e10111a7a179a3a58917a8e90e43c35f` and canonical npm tarball URL
+`https://registry.npmjs.org/owlrunkit/-/owlrunkit-0.16.0.tgz`.
+The root `owlcoda` package has a separate version lifecycle.
 
 The session is an execution interface. Project artifacts are the system of record.
 
@@ -20,13 +27,23 @@ The session is an execution interface. Project artifacts are the system of recor
 4. Never include `.owlcoda/runkit/**` in delivery changed files, profile rules or lease owned paths.
 5. Read [references/contract-v0.2.md](references/contract-v0.2.md) for new executions and accepted closeouts. Read [references/contract-v0.1.md](references/contract-v0.1.md) only when interpreting historical v0.1 artifacts.
 
-Set `SKILL_ROOT` to this skill directory before running the examples below.
+Use the `owlrunkit` command for normal work. Its bootstrap resolves the
+project's exact locally bound OwlRunKit CLI before interpreting project control
+state. The currently installed package handles bootstrap-only operations such
+as initialization, exact adoption and Core-successor migration. Do not bypass
+that resolver with a global CLI or a copied Skill Core. Project-local
+delegation requires the canonical official npm tarball URL and SHA-512 SRI,
+an installed Core identity trusted by the current bootstrap, and the same Core
+identity in project Config; any mismatch fails closed.
+Set `SKILL_ROOT` to this Skill directory only when an advanced read-only helper
+below has no CLI entry.
 
 ## Initialize and inspect
 
 ```bash
 owlrunkit init --workspace "$PROJECT_ROOT"
 owlrunkit inspect --json --workspace "$PROJECT_ROOT"
+owlrunkit status --workspace "$PROJECT_ROOT"
 ```
 
 Initialization may write only `.owlcoda/runkit/`. It must not modify business source.
@@ -65,13 +82,13 @@ repair surfaces. OwlRunKit uses its own version lifecycle: the standalone
 For one low-risk exact command, use the Core v0.13 Quick path:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" quick-verify \
+owlrunkit quick-verify \
   --workspace "$PROJECT_ROOT" -- "$ABSOLUTE_EXECUTABLE" "arg-one"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" quick-attest \
+owlrunkit quick-attest \
   --workspace "$PROJECT_ROOT" --receipt "$QUICK_RECEIPT"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" quick-metrics \
+owlrunkit quick-metrics \
   --workspace "$PROJECT_ROOT" --local
 ```
 
@@ -81,13 +98,13 @@ local-only and perform no telemetry or network request.
 Core v0.13 also adds deterministic repair and exact-byte offline transport:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" repair \
+owlrunkit repair \
   --workspace "$PROJECT_ROOT" --run-id "$RUN_ID"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" offline-export \
+owlrunkit offline-export \
   --workspace "$PROJECT_ROOT" --receipt "$RECEIPT" --output "$BUNDLE"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" offline-import \
+owlrunkit offline-import \
   --workspace "$PROJECT_ROOT" --bundle "$BUNDLE" --store "$OFFLINE_STORE"
 ```
 
@@ -111,18 +128,29 @@ node "$RUNKIT_PACKAGE_ROOT/scripts/runkit-contract/install-codex-skill.mjs" \
 
 node "$RUNKIT_PACKAGE_ROOT/scripts/runkit-contract/install-codex-skill.mjs" \
   install --repository "$RUNKIT_PACKAGE_ROOT" \
-  --target "$RUNKIT_SKILL_TARGET" --archive "$RUNKIT_SKILL_ARCHIVE_ROOT" \
-  --workspace /absolute/project-a --workspace /absolute/project-b
+  --target "$RUNKIT_SKILL_TARGET" --archive "$RUNKIT_SKILL_ARCHIVE_ROOT"
 ```
 
 The install manifest binds every Skill, Core, contract and attest file.
 Unknown installed drift fails closed instead of being overwritten. A changed
-process-wide Skill requires an explicit fleet: pass every project that consumes
-that Skill as a repeated `--workspace`. The installer validates every project
-config and refuses the whole update before replacing the Skill or changing any
-project when one project has an active execution, an active writer lease,
-ambiguous control state, or invalid configuration. Omitting the fleet is an
-error; do not pass a partial fleet to bypass this guard.
+process-wide Skill requires a registered complete fleet. Register each coverage
+root once with `owlrunkit fleet register-root --fleet-root <absolute-root>`.
+Inspect and correct a broad or stale registry through the official
+`fleet inspect-registry` and evidence-bound `fleet replace-registry` workflow;
+never hand-edit the registry. Replacement is dry-runnable and append-only
+receipts preserve the exact rollback preimage. Retired archives are evidence,
+not active fleet writers. Ordinary installs use the registered active
+membership automatically. Repeated
+`--fleet-root` or a frozen `--fleet-manifest` remain explicit one-off sources.
+RunKit stops at each project boundary, reports damaged or incomplete markers by
+exact path, freezes the discovered workspace list, and fails closed if a
+registered root is unreachable. It cannot
+discover projects outside registered coverage and must not claim machine-wide
+safety. The installer validates every discovered project config and refuses the
+whole update before replacing the Skill or changing any project when one
+project has an active execution, an active writer lease, ambiguous control
+state, or invalid configuration. A missing registry is an error; do not use
+partial explicit coverage to bypass this guard.
 
 A successful update archives the exact prior Skill, migrates each inactive
 project, and writes an append-only upgrade receipt containing the prior Skill
@@ -156,16 +184,14 @@ fleet with no matching active execution. An accepted receipt, ready-for-commit
 receipt, package publication, valid install, rollback, or recovery still does
 not grant Git, npm, release, deploy, or foreign-project authority.
 
-## Core v0.15 consolidated onboarding and compatibility
+## Core v0.14 and v0.15 historical compatibility
 
-Core v0.14 introduced self-service onboarding. The bundled Core is now
-`0.15.1`; new users should not need internal module calls,
-an external schema validator, or the full Formal lifecycle merely to understand
-whether a project can adopt RunKit.
+Core v0.14 introduced `owlrunkit doctor`, registry-first exact adoption and
+deterministic `profiles detect`; Core v0.15/0.15.1 consolidated onboarding and
+made inactive 0.13/0.14 ConfigV2 upgrades and exact rollback safe. Historical
+projects, profiles, executions, packets, receipts and closeouts remain readable.
 
 ```bash
-owlrunkit --help
-owlrunkit --version
 owlrunkit doctor --workspace "$PROJECT_ROOT"
 owlrunkit profiles detect --workspace "$PROJECT_ROOT" --dry-run
 owlrunkit profiles validate --workspace "$PROJECT_ROOT"
@@ -173,46 +199,42 @@ owlrunkit profiles impact --workspace "$PROJECT_ROOT" --changed path/to/file
 owlrunkit adopt --workspace "$PROJECT_ROOT" --exact owlrunkit@0.15.1
 ```
 
-`doctor` and `profiles detect --dry-run` are read-only. Registry diagnostics use
-a short timeout and distinguish offline or DNS/timeout failures, authorization
-failures, missing versions and local-install binding mismatches without
-printing credentials. Profile detection is deterministic evidence for a human
-decision; it does not write or overwrite project configuration.
-
-`adopt` accepts only an exact `owlrunkit@0.15.1` binding from the official npm
-registry. It recomputes the registry and local install/lock facts, invokes the
-strict adoption evaluator, and writes typed project evidence only after an
-eligible result. Worktrees, `file:` dependencies, local tarballs, Git sources,
-workspace links and mutable tags remain ineligible.
-
-Detected commands use canonical absolute regular-file launchers, including
-NVM-managed npm installs, so a detector result does not conflict with Formal
-symlink rejection. Workspace discovery understands safe literal and one-level
-`apps/*` / `packages/*` patterns and emits exact root-owned npm workspace
-commands. Detection remains dry-run only and never rewrites an existing
-profile.
-
-`doctor` warns when explicit broad formatter, linter, or test-collector
-commands can include `.owlcoda/runkit/`; it never edits ignore or tool
-configuration. `init` and `inspect` consume valid current-version adoption and
-profile facts only while the exact local registry binding remains intact and
-every launcher is Formal-safe; otherwise launcher review or adoption remains
-explicitly incomplete. The recovery action remains authoritative. A rejected
-Formal executable preflight or non-accepted finalize gate is preserved under
-the execution before the CLI returns the result.
-
-Contract v0.2 remains unchanged. Upgrading a 0.13 or 0.14 ConfigV2 project
-is allowed only while its RunKit control state is inactive and unambiguous. An
-active execution or writer lease refuses the upgrade before `config.json` or
-migration evidence changes. A legal refresh writes an exclusive sequenced
-Core-refresh migration receipt and changes only `config.json`; a repeated
-transition cannot replace its first receipt. Compatibility gates invoke npm
-through each selected Node runtime, prove active 0.13/0.14 work blocks 0.15
-before mutation, and prove upgrade-receipt rollback restores the exact prior
-Skill manifest and ConfigV2 bytes. Existing profiles, adoption evidence,
-executions, delivery packets, snapshots, verification receipts, and closeouts
-remain byte-identical and readable. Neither compatibility nor a completed
-onboarding projection grants authority.
+Exact adoption requires the official npm registry `shasum`, `integrity` and
+tarball binding. A worktree, local tarball, `file:` dependency, Git source,
+workspace link or mutable tag is not formal adoption. Active or ambiguous work
+blocks migration before Config or Skill bytes change. Historical closed leases
+use their trusted immutable closeout/Core semantics only when an accepted
+closed execution exactly binds its released lease. Active, malformed,
+hash-wrong, untrusted, or inconsistently released state still fails closed.
+Compatibility, onboarding,
+rollback and recovery grant no Git, npm, release, deploy or foreign-write
+authority.
+
+## Core v0.16 daily workflow candidate
+
+Core 0.16 makes the safe path the ordinary path without weakening its gates:
+
+```bash
+owlrunkit fleet register-root --fleet-root /absolute/declared-projects
+owlrunkit fleet discover
+owlrunkit profiles detect --workspace "$PROJECT_ROOT" --dry-run
+owlrunkit profiles detect --workspace "$PROJECT_ROOT" --apply
+owlrunkit assurance route --workspace "$PROJECT_ROOT" \
+  --request "$ASSURANCE_REQUEST_JSON"
+owlrunkit status --workspace "$PROJECT_ROOT"
+```
+
+Fleet discovery is complete only within all registered coverage roots, an
+explicit one-off root list, or a caller-supplied frozen `--fleet-manifest`. It
+refuses unreachable roots and cannot protect projects outside registered
+coverage. `profiles detect --apply` is create-only and adopts only unambiguous
+high-confidence, project-local regular-file launchers; existing profiles and
+low-confidence candidates remain untouched. Lock-bound npm and exact
+project-installed pnpm, Yarn, and Bun launchers are supported; PATH-only
+package managers remain review-required. `assurance route`
+deterministically selects No RunKit, Quick Verification or Formal Delivery
+from explicit risk facts and fails closed to Formal when required facts are
+missing.
 
 ## Choose the smallest assurance lane
 
@@ -232,59 +254,152 @@ If work crosses lanes, re-evaluate risk and authority. Moving from Quick to
 Formal requires a fresh GoalContract, exact write scope, profiles and authority;
 never treat a Quick receipt as Formal acceptance.
 
-## Plan one execution
+## Use Formal Delivery in three commands
 
-Prepare a JSON goal contract, then run:
+For a normal formally accepted change, start one bounded execution, add
+incremental checks, then finish once:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" plan \
-  --workspace "$PROJECT_ROOT" \
-  --run-id "$RUN_ID" \
-  --goal "$GOAL_JSON"
-```
-
-Use one run id on one branch only. Assign one active lease per non-overlapping write scope. Feature work may stop at `ready_for_stage_verification` with a DeliveryPacket.
-
-## Use the high-level daily lifecycle
-
-For a normal single-command verification path, compose the same atomic Core
-artifacts with three commands:
+NODE_BIN="$(node -p 'process.execPath')"
 
-```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" start \
+owlrunkit formal start \
   --workspace "$PROJECT_ROOT" --run-id "$RUN_ID" --goal "$GOAL_JSON" \
   --work-item "$WORK_ITEM_ID" \
   --owned-path "src/exact-file.ts" --owned-path "tests/feature/**"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" verify \
+owlrunkit formal check \
   --workspace "$PROJECT_ROOT" --run-id "$RUN_ID" \
-  --from-lease "$WORK_ITEM_ID" --verification-id "$VERIFICATION_ID" \
-  --cwd "." -- "$ABSOLUTE_EXECUTABLE" "arg-one" "arg-two"
+  --from-lease "$WORK_ITEM_ID" --check-id "$CHECK_ID" \
+  --cwd "." -- "$NODE_BIN" --check "src/exact-file.ts"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" finish \
+owlrunkit formal finish \
   --workspace "$PROJECT_ROOT" --run-id "$RUN_ID" --decision accepted
 ```
 
-`start` rolls back a newly created execution if lease acquisition fails.
-`verify` preserves the exact argv after `--`, derives the DeliveryPacket,
-verification context, snapshot request and finalize request, and never
-finalizes a failed snapshot. The executable may have zero arguments and argv
-may contain empty strings; no shell normalization is allowed. High-level verify
-holds the control transaction through command completion, and its snapshot
-source map must match the same DeliveryPacket at finalize and accepted finish.
-`finish accepted` derives the unique Contract v0.2 active gate before releasing
-leases and closing; a persistence failure restores leases, and project-truth
-directories with symlink ancestors fail before mutation. `blocked` and
-`rejected` remain honest non-accepted closeouts. These are convenience
-composition commands, not a second truth model. The atomic commands below
-remain available for recovery and unusual workflows.
+`formal start` creates the execution and exact writer lease. `formal check`
+preserves exact argv and appends source-bound evidence without rebuilding the
+execution. `formal finish` performs final source-drift, evidence, permission,
+acceptance, closeout and lease-release gates. `blocked` and `rejected` remain
+honest non-accepted closeouts. Keep one execution while scope is unchanged;
+create a linked successor only when authority or write scope truly changes.
+Repeating the same `check` or `finish` command resumes exact create-only
+artifacts; a mismatch fails closed.
+
+Formal accepted evidence currently admits only the built-in Node syntax check:
+the exact `process.execPath`, `--check`, and an ordinary workspace file.
+Profile, package-manager, wrapper and custom commands are rejected before
+spawn because RunKit does not yet enforce a filesystem-write and network
+sandbox for them. Full controlled execution and legacy `verify` remain
+available, but their deny rules are not sandbox-equivalent and their receipts
+must not be silently promoted into this Formal accepted-evidence path.
+
+## Freeze a dirty SourceCandidateV2
+
+Freeze, verify and materialize an immutable candidate without requiring a Git
+commit:
+
+```bash
+owlrunkit candidate freeze --workspace "$PROJECT_ROOT" \
+  --run-id "$RUN_ID" --from-lease "$WORK_ITEM_ID" \
+  --candidate-id "$CANDIDATE_ID"
+owlrunkit candidate verify --workspace "$PROJECT_ROOT" \
+  --candidate "$SOURCE_CANDIDATE"
+owlrunkit candidate materialize --workspace "$PROJECT_ROOT" \
+  --candidate "$SOURCE_CANDIDATE" \
+  --target-workspace "$CLEAN_WORKSPACE"
+```
+
+`SourceCandidateV2` binds the baseline, exact owned-path closure, changed file
+states, payload bytes and source manifest. Materialization is create-only and
+byte-verified. Payload writes and mode changes happen in a sibling staging
+checkout; the clean target is atomically switched only after verification.
+Failures restore the original target, and an interrupted switch resumes from
+the transaction journal. It is release-candidate evidence, not Git or publish
+authority.
+
+## Upgrade a self-hosted Core with a successor
+
+Use a frozen SourceCandidateV2 and the registered fleet boundary:
+
+```bash
+owlrunkit core-successor plan --workspace "$PROJECT_ROOT" \
+  --plan-id "$PLAN_ID" --run-id "$RUN_ID" \
+  --from-lease "$WORK_ITEM_ID" --candidate-id "$CANDIDATE_ID"
+owlrunkit core-successor apply --workspace "$PROJECT_ROOT" \
+  --plan "$SUCCESSOR_PLAN" --owner-authority "$OWNER_AUTHORITY" \
+  --receipt-id "$RECEIPT_ID"
+```
+
+Planning proves complete path closure, materializes the candidate in a clean
+workspace, recomputes the proposed Core identity, and freezes the registered
+fleet. Apply rediscovers the fleet, supports mixed prior Core identities,
+requires a signed V2 Owner authority from the fixed user trust store, and writes
+append-only per-project migration receipts. Unsigned V1 authorities cannot
+execute. No execution result is silently inherited.
+
+## Use two-stage deployment and the remote adapter
+
+Separate build acceptance from permission to mutate a machine:
+
+```bash
+owlrunkit deployment prepare --workspace "$PROJECT_ROOT" \
+  --run-id "$PREPARE_RUN_ID" --artifact "$ARTIFACT" \
+  --owner-decision "$OWNER_DECISION" \
+  --media-type application/gzip --output "$PREPARE_RECEIPT"
+owlrunkit deployment execute --workspace "$PROJECT_ROOT" \
+  --prepare "$PREPARE_RECEIPT" \
+  --owner-decision "$OWNER_DECISION" \
+  --owner-authority "$OWNER_AUTHORITY" \
+  --manifest "$REMOTE_MANIFEST"
+```
+
+`deployment prepare` binds the accepted source artifact, exact build artifact,
+and independent canonical `OwnerDeploymentDecisionV1`, but grants no deploy
+permission. The decision must explicitly state deployment mode, existing
+assets, rollback and data authority, service activation, baseline cut, and
+destructive scope; unset high-risk fields fail closed. A Formal `SourceCandidateV2`,
+including ordinary modifications and deletion-only candidates, remains the
+first-class source artifact through the verification receipt, accepted
+closeout, prepare receipt, and deployment lineage. Legacy DeliveryPacket-based
+accepted runs remain compatible. `deployment execute` requires a separate active execution
+and signed V2 Owner authority binding the full remote intent and exact decision
+hash. Before child creation or lease acquisition, RunKit compiles the decision
+against Goal, prepare receipt, manifest, and authority. A conflict creates no
+executable child and sends no remote command. The standard
+remote deployment adapter binds VM identity, create-only upload, ordered
+before/after stages, exact hashes, first-install versus update mode, deletion
+allowlists and remote smoke. It revalidates execution, lease, preflight, engine,
+and lineage before every stage and records the exact failure stage. Deletion is
+rejected unless both the manifest and the authorized execute execution permit
+it.
+
+The default execute command creates and closes the deployment child, goal,
+lease, profile, resource preflight, lineage, and result. Use `--request` with
+manual lineage/output paths only as an advanced compatibility path when another
+controller already owns those exact artifacts.
+
+`deployment execute --resume` may resume only the same child under the same
+decision. An explicitly superseding decision closes the old child as blocked
+`closed_superseded`, releases its lease, preserves the prior decision snapshot,
+and reports `businessGoalIncomplete=true` plus
+`nextAllowedAction=plan_replacement_execution`. Do not treat that closeout as
+business completion or reuse the old rollback plan, remote intent, or Owner
+authority for the replacement.
+
+For `kind: "builtin_ssh"`, the privileged
+`owlrunkit-remote-helper` is a caller-provided target-VM prerequisite; the npm
+package does not install it. Bind its exact path, protocol, version and fixed
+`execute`/`reconcile` capabilities in the remote manifest.
+`identity_preflight` must return that exact descriptor before upload or any
+later side effect. Treat a mismatch as blocking. This is a built-in SSH client
+and protocol contract, not an out-of-the-box privileged VM deployment.
 
 ## Manage leases and create delivery without hand-authored JSON
 
 Acquire one bounded writer lease with repeated explicit paths:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" lease acquire \
+owlrunkit lease acquire \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --work-item "$WORK_ITEM_ID" \
@@ -295,10 +410,10 @@ node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" lease acquire \
 Inspect or release it through the Core instead of editing lease JSON:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" lease inspect \
+owlrunkit lease inspect \
   --workspace "$PROJECT_ROOT" --run-id "$RUN_ID"
 
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" lease release \
+owlrunkit lease release \
   --workspace "$PROJECT_ROOT" --run-id "$RUN_ID" \
   --work-item "$WORK_ITEM_ID"
 ```
@@ -306,7 +421,7 @@ node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" lease release \
 After implementation, derive the bounded DeliveryPacket from the active lease:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" delivery create \
+owlrunkit delivery create \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --from-lease "$WORK_ITEM_ID" \
@@ -342,7 +457,7 @@ If any path is uncovered, use the declared full profile. Do not interpret an emp
 Before reusing evidence, prepare `OwlCodaRunKitCoverageAdoptRequestV1` and run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" coverage-adopt \
+owlrunkit coverage-adopt \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$COVERAGE_ADOPT_REQUEST_JSON"
@@ -362,7 +477,7 @@ After inspecting the source execution, prepare `OwlCodaRunKitResumeRequestV1`
 and run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" resume \
+owlrunkit resume \
   --workspace "$PROJECT_ROOT" \
   --run-id "$SOURCE_RUN_ID" \
   --request "$RESUME_REQUEST_JSON"
@@ -392,7 +507,7 @@ After freezing a DeliveryPacket, prepare
 SHA-256, then run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" verify-plan \
+owlrunkit verify-plan \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$VERIFY_PLAN_REQUEST_JSON"
@@ -423,7 +538,7 @@ For a model workload, first bind the remaining work to a hashed
 `OwlCodaRunKitResourcePreflightRequestV1` and run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" resource-preflight \
+owlrunkit resource-preflight \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$RESOURCE_PREFLIGHT_REQUEST_JSON"
@@ -468,7 +583,7 @@ labels are not acceptance evidence.
 Prepare `OwlCodaRunKitSnapshotRequestV1`, then run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" snapshot \
+owlrunkit snapshot \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$SNAPSHOT_REQUEST_JSON"
@@ -490,7 +605,7 @@ and after the command.
 Prepare `OwlCodaRunKitVisualSmokeRequestV1`, then run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" visual-smoke \
+owlrunkit visual-smoke \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$VISUAL_SMOKE_REQUEST_JSON"
@@ -510,7 +625,7 @@ After every required snapshot passed and the DeliveryPacket is still fresh,
 prepare `OwlCodaRunKitFinalizeRequestV1` and run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" finalize \
+owlrunkit finalize \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$FINALIZE_REQUEST_JSON"
@@ -527,7 +642,7 @@ this command is available.
 Close only after consuming the current stage receipt and making an explicit controller decision:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" closeout \
+owlrunkit closeout \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --decision accepted \
@@ -542,20 +657,24 @@ passing gate so failed work still leaves durable truth.
 
 ## Produce ready-for-commit evidence
 
-After an accepted closeout, capture fresh project-root snapshots, prepare
-`OwlCodaRunKitReadyForCommitRequestV1`, then run:
+After an accepted closeout, capture fresh project-root snapshots and start from
+`assets/templates/ready-for-commit-request.json`. The default
+`OwlCodaRunKitReadyForCommitRequestV2` binds exactly one
+`sourceCandidatePath` or legacy `deliveryPacketPath`, then run:
 
 ```bash
-node "$SKILL_ROOT/scripts/runkit-contract/runkit-cli.mjs" ready-for-commit \
+owlrunkit ready-for-commit \
   --workspace "$PROJECT_ROOT" \
   --run-id "$RUN_ID" \
   --request "$READY_REQUEST_JSON"
 ```
 
-The command revalidates the DeliveryPacket, profile impact, gate input,
-accepted closeout, released leases, and selected root manifests. It writes
-`READY_FOR_COMMIT_RECEIPT.json` with every repository action and
-`authorizationGranted` set to false. A ready receipt records state; it does
+The command revalidates the selected source artifact, profile impact, gate
+input, accepted closeout, released leases, and selected root manifests.
+SourceCandidateV2 remains exact across additions, modifications, deletions, and
+renames. V1 DeliveryPacket requests remain supported for historical executions.
+The command writes `READY_FOR_COMMIT_RECEIPT.json` with every repository action
+and `authorizationGranted` set to false. A ready receipt records state; it does
 not authorize Git.
 
 ## Foreign-project shadow

package/integrations/codex/skills/owlcoda-runkit/assets/templates/config.json

@@ -2,9 +2,9 @@
   "schemaVersion": "OwlCodaRunKitConfigV2",
   "core": {
     "contractVersion": "0.2",
-    "coreVersion": "0.15.1",
-    "coreManifestSha256": "sha256:replace-with-installed-core-manifest",
-    "coreSourceRef": "artifact:sha256:replace-with-content-addressed-source"
+    "coreVersion": "0.16.1",
+    "coreManifestSha256": "sha256:38ac9110e328c38a81db05f1359734ca810b991c9f4de77647f9719f5e6af78b",
+    "coreSourceRef": "artifact:sha256:38ac9110e328c38a81db05f1359734ca810b991c9f4de77647f9719f5e6af78b"
   },
   "authorizationPolicy": "external_explicit_authority_required"
 }

package/integrations/codex/skills/owlcoda-runkit/assets/templates/ready-for-commit-request.json

@@ -1,6 +1,6 @@
 {
-  "schemaVersion": "OwlCodaRunKitReadyForCommitRequestV1",
-  "deliveryPacketPath": ".owlcoda/runkit/executions/replace-with-run-id/delivery-packets/replace-with-packet.json",
+  "schemaVersion": "OwlCodaRunKitReadyForCommitRequestV2",
+  "sourceCandidatePath": ".owlcoda/runkit/executions/replace-with-run-id/source-candidates/replace-with-candidate.json",
   "verificationGateInputPath": ".owlcoda/runkit/executions/replace-with-run-id/verification-receipts/replace-with-receipt/verification-gate-input.json",
   "roots": [
     {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "owlrunkit",
-  "version": "0.15.1",
+  "version": "0.16.1",
   "description": "Project-owned execution, evidence, verification, and recovery for coding agents.",
   "license": "GPL-3.0-or-later",
   "author": "yeemio <owlclaw@163.com>",
@@ -11,7 +11,7 @@
     "registry": "https://registry.npmjs.org"
   },
   "bin": {
-    "owlrunkit": "scripts/runkit-contract/runkit-cli.mjs",
+    "owlrunkit": "scripts/runkit-contract/runkit-bootstrap.mjs",
     "owlrunkit-attest": "packages/attest/cli/owlcoda-attest.mjs"
   },
   "exports": {

package/packages/attest/distribution-manifest-v1.json

@@ -36,6 +36,16 @@
     "sha256": "sha256:3972dc9744f6499f0f9b2dbf76696f2ae7ad8af9b23dde66d6af86c9dfb36986"
   },
   "supportedProducerCoreIdentities": [
+    {
+      "contractVersion": "0.2",
+      "coreVersion": "0.16.1",
+      "coreManifestSha256": "sha256:38ac9110e328c38a81db05f1359734ca810b991c9f4de77647f9719f5e6af78b"
+    },
+    {
+      "contractVersion": "0.2",
+      "coreVersion": "0.16.0",
+      "coreManifestSha256": "sha256:320fb1d97b4459d1a14b0b67807dcf2bc6b03970492cb9fb2c6245b17912c81e"
+    },
     {
       "contractVersion": "0.2",
       "coreVersion": "0.15.1",

package/packages/attest/src/quick.mjs

@@ -18,6 +18,16 @@ import { captureWorkspaceSnapshot } from "./workspace-snapshot.mjs";
 export { validQuickReceiptShape } from "./quick-receipt-contract.mjs";
 
 export const SUPPORTED_QUICK_CORE_IDENTITIES = Object.freeze([
+  {
+    contractVersion: "0.2",
+    coreVersion: "0.16.1",
+    coreManifestSha256: "sha256:38ac9110e328c38a81db05f1359734ca810b991c9f4de77647f9719f5e6af78b",
+  },
+  {
+    contractVersion: "0.2",
+    coreVersion: "0.16.0",
+    coreManifestSha256: "sha256:320fb1d97b4459d1a14b0b67807dcf2bc6b03970492cb9fb2c6245b17912c81e",
+  },
   {
     contractVersion: "0.2",
     coreVersion: "0.15.1",

package/registry-release-policy-v1.json

@@ -2,7 +2,7 @@
   "schemaVersion": "OwlCodaRunKitRegistryReleasePolicyV1",
   "status": "release_candidate_unpublished",
   "packageName": "owlrunkit",
-  "candidateVersion": "0.15.1",
+  "candidateVersion": "0.16.1",
   "registry": "https://registry.npmjs.org",
   "registryRequiredBeforeAdoption": true,
   "exactVersionRequired": true,
@@ -21,7 +21,13 @@
     "symlink",
     "workspace"
   ],
-  "currentPublishedRelease": null,
+  "currentPublishedRelease": {
+    "packageName": "owlrunkit",
+    "version": "0.16.0",
+    "shasum": "bab651b0e10111a7a179a3a58917a8e90e43c35f",
+    "integrity": "sha512-qrlJueVypcetnFXAEHhDcYlZvpJAnTsbtLu9Rtp9SXNekAAD/rtBM8p2GqM5dOc1ERojixc3ipQkDGw6T7IR1A==",
+    "tarballUrl": "https://registry.npmjs.org/owlrunkit/-/owlrunkit-0.16.0.tgz"
+  },
   "registryRelease": null,
   "registryEvidenceMustBeExternal": true,
   "registryEvidenceSchemaVersion": "OwlCodaRunKitRegistryReleaseEvidenceV1",