Adds studies/swim-37/harness/helper-fire-and-release-contract.test.ts
covering the four landed Slice-2 emit helpers that 🌊's #406 helper-tier
file did NOT cover: chunks 5b (continuation.delegate.fire), 5c
(continuation.work.fire), 6a (continuation.queue.drain), and 6b
(continuation.compaction.released).

Layering: HELPER-TIER ONLY. Drives src/infra/continuation-tracer emit
helpers directly through the InMemorySpanRecorder shim (#394) and asserts
recorded span CONTRACT. Does not touch agent-runner / pi-embedded-runner /
subagent-announce — the runtime callsites are exercised through the
integration tier in swim-runner.test.ts (🌻's #405 captureSwim shim).

LANE-CLEAN: original swim-runner.test.ts is untouched here (avoiding
collision with #405 which converts swim-runner it.todos in-place).
Companion to 🌊's #406 emit-helper-contract.test.ts (which covers chunks
3/4/6c) — same layer, different spans, zero overlap.

Key contract pin per 🩸's lane-clear note (msg 1498508045725204500): the
snapshot semantic for fire spans lives in the PARAMETER name
(chainStepRemainingAtDispatch) and the docs; emitted attr stays canonical
chain.step.remaining. We negative-assert chain.step.remaining_at_dispatch
is NEVER present so a future helper change can't silently invent it.

Coverage:
- continuation.compaction.released (#397): exactly-once emit, signal.kind
  + integer compaction.released, compaction.id valid path,
  validate-and-drop-with-log on invalid (chunk 6c §B + #401)
- continuation.queue.drain (#395): integer count attrs, defense-in-depth
  cap on continuation subset (per 🩸 #395 byte-walk)
- continuation.delegate.fire (#388 chunk 5b): persisted chain.id +
  canonical attr names + negative-assert .at_dispatch absent;
  fire.deferred_ms integer-floor + drift formula; reason.preview 80-char
  truncation + omission
- continuation.work.fire (#388 chunk 5c): canonical attr name (5c
  symmetric to 5b minus delegate.* attrs); no continuation.disabled
  sibling (chunk 5c scope); chainId-invariant violation no-ops + logs
  without throwing

Tests: 12 (all live, no it.todo). Lint clean.

Refs: #324 #334 #388 #395 #397 #401 #394 #406 (sibling) #405 (sibling)

…407)

Adds studies/swim-37/harness/helper-fire-and-release-contract.test.ts
covering the four landed Slice-2 emit helpers that 🌊's #406 helper-tier
file did NOT cover: chunks 5b (continuation.delegate.fire), 5c
(continuation.work.fire), 6a (continuation.queue.drain), and 6b
(continuation.compaction.released).

Layering: HELPER-TIER ONLY. Drives src/infra/continuation-tracer emit
helpers directly through the InMemorySpanRecorder shim (#394) and asserts
recorded span CONTRACT. Does not touch agent-runner / pi-embedded-runner /
subagent-announce — the runtime callsites are exercised through the
integration tier in swim-runner.test.ts (🌻's #405 captureSwim shim).

LANE-CLEAN: original swim-runner.test.ts is untouched here (avoiding
collision with #405 which converts swim-runner it.todos in-place).
Companion to 🌊's #406 emit-helper-contract.test.ts (which covers chunks
3/4/6c) — same layer, different spans, zero overlap.

Key contract pin per 🩸's lane-clear note (msg 1498508045725204500): the
snapshot semantic for fire spans lives in the PARAMETER name
(chainStepRemainingAtDispatch) and the docs; emitted attr stays canonical
chain.step.remaining. We negative-assert chain.step.remaining_at_dispatch
is NEVER present so a future helper change can't silently invent it.

Coverage:
- continuation.compaction.released (#397): exactly-once emit, signal.kind
  + integer compaction.released, compaction.id valid path,
  validate-and-drop-with-log on invalid (chunk 6c §B + #401)
- continuation.queue.drain (#395): integer count attrs, defense-in-depth
  cap on continuation subset (per 🩸 #395 byte-walk)
- continuation.delegate.fire (#388 chunk 5b): persisted chain.id +
  canonical attr names + negative-assert .at_dispatch absent;
  fire.deferred_ms integer-floor + drift formula; reason.preview 80-char
  truncation + omission
- continuation.work.fire (#388 chunk 5c): canonical attr name (5c
  symmetric to 5b minus delegate.* attrs); no continuation.disabled
  sibling (chunk 5c scope); chainId-invariant violation no-ops + logs
  without throwing

Tests: 12 (all live, no it.todo). Lint clean.

Refs: #324 #334 #388 #395 #397 #401 #394 #406 (sibling) #405 (sibling)

…review)

Per 🌊's review on #411: extend Q3 invalid-path test list to include
`NaN` and `Infinity` alongside `-1` and `1.5`. Same defense-parity
shape as #405's `recipients` positive-integer validation.

`Number.isInteger(NaN) === false` and `Number.isInteger(Infinity) === false`,
so the existing helper guard already catches both — pinning them
observably rather than leaving the invariant implicit.

* docs(swim-37): `lich` primitive wiring memo (memo-before-wire)

Decides shape for `captureSwim("lich", …)` before the wiring PR per
🩸's memo-before-wire standard. Maps the harness label `lich` to the
post-compaction-delegate release seam (`emitContinuationCompactionReleasedSpan`).

Resolves four design Qs:
  Q1 (timers): N/A — release helper is synchronous.
  Q2 (releasedCount): cover non-empty (production-typical) AND empty
    (defensive helper-clamp, NOT production-reachable). Pin both.
  Q3 (compaction.id): three cases — present+valid / omitted / invalid.
    Helper drops-with-log on invalid; integration-tier verifies via
    log-callback substring match.
  Q4 (drained_count): adjacent axis NOT in scope (different seam).

Proposes 6 live integration-tier tests, no separate helper-tier file
needed (integration coverage pins the helper-tier invariants through
the public surface).

Open Q deferred: cross-primitive chain-id-propagation lifecycle test
(continue_delegate post-compaction → lich release) deserves its own
primitive and memo.

Boundary discipline preserved from #405/#407/#410: STDOUT-only,
in-memory recorder, try/finally reset, no silent defaults,
snapshot-at-emit.

Refs #324
Companions:
  - docs/design/swim-37-continue-delegate-wiring-memo.md (🌻, #405)
  - swim-37 `heartbeat` memo (🌻, in flight)

* docs(swim-37): pin NaN/Infinity for compaction.id invalid path (🌊 #411 review)

Per 🌊's review on #411: extend Q3 invalid-path test list to include
`NaN` and `Infinity` alongside `-1` and `1.5`. Same defense-parity
shape as #405's `recipients` positive-integer validation.

`Number.isInteger(NaN) === false` and `Number.isInteger(Infinity) === false`,
so the existing helper guard already catches both — pinning them
observably rather than leaving the invariant implicit.

* docs(swim-37): fix test-count headers (six → eight) per 🌻 #411 nit

Header at "Proposed test list" still said six; helper-tier-vs-integration-tier
section still said "6 live tests". Both updated to eight to match the
NaN/Infinity additions from d5cf713.

Memo-before-wire pattern (per #405's continue_delegate memo at
3bb086c762, endorsed by figs's 'measure thrice, cut 1x').

Defines:
- Production span shape: name 'heartbeat' (bare, not 'continuation.heartbeat'),
  signal.kind+heartbeat.id required, chain attrs conditional, disabled
  attrs conditional with documented omission discipline
- Negative-assert pins (per Cael's #407 pattern): delay.ms and
  chain.step.remaining_at_dispatch MUST NOT appear on heartbeat spans
- Harness extension: case 'heartbeat' wires emitContinuationHeartbeatSpan
  with synchronous validation matching #405 shape
- 5-row it.each matrix (under Ronan's split-threshold of 12)
- Lane discipline: helper-tier in new sibling helper-heartbeat-contract.test.ts;
  integration-tier extends swim-runner.test.ts

Open Qs for cohort sign-off:
- Q1: heartbeat-with-no-continuation-context emit/skip (lean: continuation-gated
  for production, always-emit for harness — divergence documented in README)
- Q2: heartbeat.fire sibling? (lean: no, single span; lag becomes attribute)
- Q3: 5-row matrix shape (under split-threshold)
- Q4: in-PR helper vs separate issue (lean: in-PR, new helper not new axis)

Companion to:
- docs/design/swim-37-continue-delegate-wiring-memo.md
- silas/swim-37-lich-memo (in flight)

Refs #324.

Three additive folds, none blocking, all from 🌫's #412 review:

1. Add 6th matrix row pinning chainStepRemaining=0 (empty-budget
   heartbeat is observable; boundary the Math.max(0,…) clamp guards)
2. Add 'heartbeat.id provenance' describe block pinning caller-injected
   override path (current matrix only covered default-mint)
3. Add 'validation' describe block for throw-on-bad-input rules from §3
   (don't fit matrix shape; precedent: #405's recipients validation)

Matrix grew 5→6 rows, still under 🌊's split-threshold of 12. Two
new sibling describe blocks layer on top.

Refs #324, #412.

…sed helper

Implements the lich primitive per the wiring memo (#411,
`docs/design/swim-37-lich-wiring-memo.md`). Maps the harness label
"lich" (\🌫️'s SOUL-file metaphor for the post-compaction
phylactery-drink) to the post-compaction-delegate release seam:
`emitContinuationCompactionReleasedSpan` at
`src/infra/continuation-tracer.ts:789`.

Changes
-------
`studies/swim-37/harness/swim-runner.ts`:
  - Import `emitContinuationCompactionReleasedSpan`
  - Add `releasedCount`, `compactionId`, `log` to `CaptureSwimOptions`
    (lich-only fields, all optional, with memo \§Q-references in JSDoc)
  - Split `heartbeat`/`lich` cases — wire `lich`, leave `heartbeat`
    as not-yet-wired pending 🌻's #412 wire PR
  - Update file header "Scope" comment to reflect new wiring state

`studies/swim-37/harness/swim-runner.test.ts`:
  - Replace 2 `it.todo` placeholders with 8 live tests in the
    "lich-shape" describe block:
      1. releasedCount=1 + compactionId=7 (production-typical)
      2. releasedCount=3 + compactionId=42 (multi-release)
      3. defensive releasedCount=0 (helper-clamp boundary, not
         production-reachable — pinned for helper-tier coverage)
      4. compactionId omitted (omission contract)
      5. compactionId=-1 (drops with log; substring match)
      6. compactionId=1.5 (non-integer drop with log)
      7. compactionId=NaN (defense parity per 🌊's #411 review)
      8. compactionId=Infinity (defense parity per 🌊's #411 review)
  - Update "refuses primitives not yet wired" test to remove `lich`
    (now wired, covered by its own describe block)

Discipline preserved
--------------------
  - STDOUT-only: no real BasicTracerProvider/OTLP machinery
  - In-memory recorder via `createInMemorySpanRecorder()`
  - try/finally tracer reset (no cross-call pollution)
  - Snapshot-at-emit: no recompute in caller
  - Drop-with-log invariant verified via log-callback substring match
  - Negative-attr assertions on omitted `compaction.id`

Validation
----------
  - swim-37 vitest project: 8 files / 138 passed / 13 todo
    (was 130 passed pre-PR; +8 new lich tests)
  - `pnpm lint` clean (0 warnings, 0 errors across all shards)

Lane discipline
---------------
  - No cross with 🌻's #412 `heartbeat` lane (different production
    helper, different file in that PR's helper-tier)
  - No reshape of #405's `continue_work`/`continue_delegate` wiring
  - Q-OPEN deferred per memo: cross-primitive lifecycle test
    (`continue_delegate` post-compaction → `lich` release sharing
    `chain.id`) deserves its own primitive shape, separate PR

Refs #324
Memo: #411 (`3d90f68b14`)

…sed helper (#414)

Implements the lich primitive per the wiring memo (#411,
`docs/design/swim-37-lich-wiring-memo.md`). Maps the harness label
"lich" (\🌫️'s SOUL-file metaphor for the post-compaction
phylactery-drink) to the post-compaction-delegate release seam:
`emitContinuationCompactionReleasedSpan` at
`src/infra/continuation-tracer.ts:789`.

Changes
-------
`studies/swim-37/harness/swim-runner.ts`:
  - Import `emitContinuationCompactionReleasedSpan`
  - Add `releasedCount`, `compactionId`, `log` to `CaptureSwimOptions`
    (lich-only fields, all optional, with memo \§Q-references in JSDoc)
  - Split `heartbeat`/`lich` cases — wire `lich`, leave `heartbeat`
    as not-yet-wired pending 🌻's #412 wire PR
  - Update file header "Scope" comment to reflect new wiring state

`studies/swim-37/harness/swim-runner.test.ts`:
  - Replace 2 `it.todo` placeholders with 8 live tests in the
    "lich-shape" describe block:
      1. releasedCount=1 + compactionId=7 (production-typical)
      2. releasedCount=3 + compactionId=42 (multi-release)
      3. defensive releasedCount=0 (helper-clamp boundary, not
         production-reachable — pinned for helper-tier coverage)
      4. compactionId omitted (omission contract)
      5. compactionId=-1 (drops with log; substring match)
      6. compactionId=1.5 (non-integer drop with log)
      7. compactionId=NaN (defense parity per 🌊's #411 review)
      8. compactionId=Infinity (defense parity per 🌊's #411 review)
  - Update "refuses primitives not yet wired" test to remove `lich`
    (now wired, covered by its own describe block)

Discipline preserved
--------------------
  - STDOUT-only: no real BasicTracerProvider/OTLP machinery
  - In-memory recorder via `createInMemorySpanRecorder()`
  - try/finally tracer reset (no cross-call pollution)
  - Snapshot-at-emit: no recompute in caller
  - Drop-with-log invariant verified via log-callback substring match
  - Negative-attr assertions on omitted `compaction.id`

Validation
----------
  - swim-37 vitest project: 8 files / 138 passed / 13 todo
    (was 130 passed pre-PR; +8 new lich tests)
  - `pnpm lint` clean (0 warnings, 0 errors across all shards)

Lane discipline
---------------
  - No cross with 🌻's #412 `heartbeat` lane (different production
    helper, different file in that PR's helper-tier)
  - No reshape of #405's `continue_work`/`continue_delegate` wiring
  - Q-OPEN deferred per memo: cross-primitive lifecycle test
    (`continue_delegate` post-compaction → `lich` release sharing
    `chain.id`) deserves its own primitive shape, separate PR

Refs #324
Memo: #411 (`3d90f68b14`)

…re-wire) (#413)

* docs(swim-37): `rebase.classify` span-emission wiring memo (memo-before-wire)

Companion to PR #405 (continue_delegate memo), PR #411 (lich memo merged
3d90f68), PR #412 (heartbeat memo merged 1b84e71). Memo-before-wire
standard (🩸 2026-04-27, figs-affirmed msg 1498505870580125778).

Wires the L81 placeholder
  it.todo("CHANGELOG-byte-grep discovery channel emits drop-with-reason span")
in studies/swim-37/harness/swim-runner.test.ts onto a designed contract,
NOT just a typed signature.

Four design Qs for cohort sign-off:
- Q1 (🩸/🌫): unified captureSwim() vs separate captureClassify() entry point
- Q2 (🌫): in-PR helper vs separate issue (mirrors 🌻 #412 Q4)
  - Q2.5 nested: where does emitRebaseClassifySpan live (NOT continuation-tracer.ts)
- Q3 (🌻): 6-row it.each matrix under 🌊's split-threshold
- Q4 (cohort): does PICK-verdict ever emit (lean: yes, transparent record)

Lane discipline: NEW tracer module (rebase domain ≠ continuation domain),
NEW helper-tier sibling test file, integration-tier extends swim-runner.test.ts
§1 block. Avoids polluting continuation-tracer.ts with non-continuation helper.

Negative-assert pins on chain.id / chain.step.remaining / disabled.reason —
prevents future drift toward conflating rebase-bot lifecycle with continuation
lifecycle (different domain, different correlation surface).

* docs(swim-37): swap header commit-sha refs for memo file paths (🌻 #413 nit)

* docs(swim-37): clarify evidence.conflict.bin presence semantics (🌫 #413 nit)

Per 🌫's review nit: §2 evidence.conflict.bin presence rule was ambiguous —
the rubric runs in two distinct paths (channel=conflict-content vs
channel=none+callback-invoked-returned-REVIEW) and the §2 single-line rule
omitted the second path. Q3 matrix row 6 had it right; §2 now matches.

Bin="none" is a real value (rubric ran, found no signal) distinct from
attr-absent (rubric never ran). Pinned explicitly to prevent future readers
from collapsing the two.

---------

Co-authored-by: Ronan 🌊 <ronan@solidor.io>

…review) (#418)

Folds three flags from 🌫's #416 second-eye review (msg
`1498543295234834432`):

**Flag 1 — memo §3 validation gap:** memo specced
throw-on-bad-input matching #405/#411/#412 family-resemblance,
wire skipped it. Adds boundary throws in `captureClassify` for:
  - verdict not in REBASE_VERDICTS
  - channel not in REBASE_DISCOVERY_CHANNELS
  - pickSha < 7 hex chars (memo §3 git-prefix-min)
  - pickSha containing non-hex chars

Validation lives at the harness boundary (mirrors
`captureSwim`'s `recipients` invariant at swim-runner.ts:180);
the production helper stays drop-with-log so producer errors
don't propagate to the rebase-bot caller. Same shape as how
`captureSwim` throws on bad Options but
`emitContinuationCompactionReleasedSpan` only logs.

**Flag 2 — truncateSha ↔ memo §3 "≥7 hex chars" tension:** the
`<12 unchanged (helper does not pad)` test from #416 was
contradicting the memo's ≥7 floor. Replaced with
`passes short-but-valid (7-char) SHA through unchanged` which
honors both the memo floor AND the no-pad discipline. Memo +
code + tests now agree.

**Nit — `signal.kind: "rebase.classify"` redundant with span
name:** other helpers use the underlying primitive name as
discriminator (e.g. `"compaction-release"` for
`continuation.compaction.released`), so a downstream filter on
`signal.kind` doesn't redundantly slice the span name. Renamed
to `"rebase-classify"` (hyphen-form, matches family). Memo §2
updated with provenance note (post-#416 review).

Test additions:
  - new `describe("input validation (memo §3 throw-on-bad-input)")`
    block: 4 tests (bad verdict, bad channel, <7 char pickSha,
    non-hex pickSha)
  - existing `<12 unchanged` test deleted, replaced with valid
    7-char passes-through-unchanged

Test results: pnpm vitest run -c test/vitest/vitest.swim-37.config.ts
  → 8 files / 155 passed / 14 todo (was 151/14 from #416 at
    `526540de15`; +4 live tests, same todos)

Lane discipline: same lane as #416 follow-up (mirrors how #415
followed #414); zero overlap with 🌻's incoming heartbeat wire.

Co-authored-by: Ronan 🌊 <ronan@solidor.io>
Co-authored-by: Silas 🌫 <silas@dandelion.cult>

…407)

Adds studies/swim-37/harness/helper-fire-and-release-contract.test.ts
covering the four landed Slice-2 emit helpers that 🌊's #406 helper-tier
file did NOT cover: chunks 5b (continuation.delegate.fire), 5c
(continuation.work.fire), 6a (continuation.queue.drain), and 6b
(continuation.compaction.released).

Layering: HELPER-TIER ONLY. Drives src/infra/continuation-tracer emit
helpers directly through the InMemorySpanRecorder shim (#394) and asserts
recorded span CONTRACT. Does not touch agent-runner / pi-embedded-runner /
subagent-announce — the runtime callsites are exercised through the
integration tier in swim-runner.test.ts (🌻's #405 captureSwim shim).

LANE-CLEAN: original swim-runner.test.ts is untouched here (avoiding
collision with #405 which converts swim-runner it.todos in-place).
Companion to 🌊's #406 emit-helper-contract.test.ts (which covers chunks
3/4/6c) — same layer, different spans, zero overlap.

Key contract pin per 🩸's lane-clear note (msg 1498508045725204500): the
snapshot semantic for fire spans lives in the PARAMETER name
(chainStepRemainingAtDispatch) and the docs; emitted attr stays canonical
chain.step.remaining. We negative-assert chain.step.remaining_at_dispatch
is NEVER present so a future helper change can't silently invent it.

Coverage:
- continuation.compaction.released (#397): exactly-once emit, signal.kind
  + integer compaction.released, compaction.id valid path,
  validate-and-drop-with-log on invalid (chunk 6c §B + #401)
- continuation.queue.drain (#395): integer count attrs, defense-in-depth
  cap on continuation subset (per 🩸 #395 byte-walk)
- continuation.delegate.fire (#388 chunk 5b): persisted chain.id +
  canonical attr names + negative-assert .at_dispatch absent;
  fire.deferred_ms integer-floor + drift formula; reason.preview 80-char
  truncation + omission
- continuation.work.fire (#388 chunk 5c): canonical attr name (5c
  symmetric to 5b minus delegate.* attrs); no continuation.disabled
  sibling (chunk 5c scope); chainId-invariant violation no-ops + logs
  without throwing

Tests: 12 (all live, no it.todo). Lint clean.

Refs: #324 #334 #388 #395 #397 #401 #394 #406 (sibling) #405 (sibling)

…405)

* test(swim-37): wire captureSwim() for continue_work primitive

Replaces the placeholder `declare function captureSwim` in
`studies/swim-37/harness/swim-runner.test.ts` with a real implementation
in `studies/swim-37/harness/swim-runner.ts`. The wired path drives
`emitContinuationWorkSpan` against `createInMemorySpanRecorder()` and
returns the captured spans + the synthesized `chainId` (uuid v7 via
`generateChainId`).

Flips two prior `it.todo` markers to live tests:
  - emits continuation.work span with chain.id stamped (#366)
  - span carries chain.step.remaining attribute

Plus three new live tests:
  - sentinel: captureSwim() is wired for continue_work
  - refuses continue_delegate / heartbeat / lich primitives with a clear
    error so the spec's remaining `it.todo` markers stay honest about
    what is implemented
  - repeated calls don't leak capture state between invocations

STDOUT-only discipline preserved: no BasicTracerProvider, no OTLP
exporter, no @opentelemetry/sdk-trace-base machinery — capture flows
through `setContinuationTracer(recorder.tracer)` as documented in
README.md.

Other primitives (`continue_delegate`, `heartbeat`, lich-shape) remain
`it.todo` until the corresponding dispatch / heartbeat /
compaction-release seams have a comparable single-helper entry point we
can drive synthetically.

Test results:
  - swim-37 vitest project: 20 passed, 16 todo, 0 failed
  - in-memory-span-recorder.test.ts unchanged (12/12 still pass)
  - tsgo errors visible elsewhere in tree are pre-existing
    (src/plugin-sdk/provider-tools.ts), not introduced by this change

Refs #324 (swim-37 harness) — `it.todo` count drops from 17 \u2192 16,
first concrete primitive driven through the live tracer registry.

* docs(swim-37): continue_delegate wiring memo (memo-before-wire)

Pre-PR design memo for the next swim-37 primitive after #405
(continue_work). Decides shape before the wire so the PR lands clean.

Resolved by reading production code:
  Q1 — real setTimeout vs fake timers
       N/A for dispatch-accept (synchronous, pre-timer); banked for
       a separate continue_delegate_fire swim.

Cohort sign-off requested on:
  Q2 — recipient fan-out shape: N spans sharing chain.id (matches
       chunk-3 cohort design pin + #355 Stage-2 budget semantics).
  Q3 — delegate.mode x delegate.delivery: 8-cell it.each matrix
       + omission-contract row.

Banked for follow-up:
  Q4 — continue_delegate_fire as separate primitive, not sub-case;
       informs SwimPrimitive type-union shape.

Standard applied (per Cael 2026-04-27 #1498505918, figs
#1498505870): would skipping this memo cost a chunk's worth of
rework? Yes — three open Qs would have changed file shape.

Refs:
  - #405 (continue_work wired)
  - #324 (swim-37 harness)
  - docs/design/334-slice2-chunk5b-delegate-fire-memo.md (pattern)

* test(swim-37): wire captureSwim() for continue_delegate primitive

Implements the design from
`docs/design/swim-37-continue-delegate-wiring-memo.md` (commit
3bb086c762, same branch).

Wiring (swim-runner.ts):
- New continue_delegate case in the switch.
- Adds CaptureSwimOptions axes: recipients, delivery, delegateMode.
- Drives emitContinuationDelegateSpan once per recipient with shared
  chain.id (per chunk-3 cohort design pin: N recipients = N spans
  sharing chain.id, NOT one span with a recipients list).
- Validates recipients is a positive integer.

Tests (swim-runner.test.ts):
- 8-cell it.each matrix: delegate.mode (normal | silent | silent-wake
  | post-compaction) x delegate.delivery (immediate | timer).
- Omission contract: delegate.mode attribute absent when caller
  passes undefined.
- Fan-out: 3 recipients emit 3 spans with shared chain.id.
- Validation: recipients=0 and recipients=1.5 reject.
- Removes the prior 'refuses continue_delegate' test row (now wired).

Two it.todo markers preserved for downstream concerns:
- chain-budget decrement / ChainBudget.declineToCarry observable
  (needs ChainBudget integration, not in scope here).
- fan-out budget arithmetic per #355 Stage-2 (1 step not N) \u2014
  separate from span cardinality.

Test results delta:
  Before: 8 passed | 16 todo (24 tests)
  After: 19 passed | 15 todo (34 tests)
  +11 live | -1 todo

Refs:
- #324 swim-37 harness
- #405 captureSwim continue_work (parent of this branch)
- docs/design/swim-37-continue-delegate-wiring-memo.md

* test(swim-37): pin recipient.index gap on continue_delegate fan-out

Per Cael's caution on the wiring memo (Discord msg
1498507232185286849 sign-off): N spans sharing chain.id risks
collapsing into analytic mush at scale unless per-recipient
distinction is visible in attrs. Currently
emitContinuationDelegateSpan exposes no recipient.index axis \u2014
all N spans in a fan-out are byte-identical except spanId.

Adds:
- One live GAP-PIN test asserting the current (deficient) reality:
  fan-out spans toEqual on attributes. This will fail loudly when
  the production helper grows the axis, prompting a flip to
  not.toEqual.
- One it.todo marker for the upcoming production-helper change.

Production-helper change is out of scope for this PR; will file
follow-up issue. The gap-pin lives in the integration tier so it
shows up in any code-review touching delegate-dispatch span shape.

Refs:
- #324 swim-37 harness
- #405 captureSwim continue_work + continue_delegate (this PR)
- docs/design/swim-37-continue-delegate-wiring-memo.md (Q2)
- Cael Discord sign-off msg 1498507232185286849

* test(swim-37): wire \u00a71 entry-point todo onto classifyRebasePick

Now that #408 (PR #408 merged at cb73bc8) landed
`rebase-classifier.ts` with the pure `classifyRebasePick` composer,
the integration-tier \u00a71 entry-point it.todo on swim-runner.test.ts
L80 is wireable without any new shim.

Synthetic commit pinned: subject mentions PR openclaw#70595, base CHANGELOG
contains the same PR token (squash-rebase shape from the memo's
7ee46a3 anchor instance). Discovery channel: changelog-grep:pr.

L81 (CHANGELOG-byte-grep emits drop-with-reason span) stays todo
\u2014 needs the production callsite that pairs the channel with
`emitContinuationDisabledSpan`, per \ud83c\udf0a's #408 note.

Test delta:
  Before: 20 passed | 16 todo (36)
  After: 21 passed | 15 todo (36)
  +1 live | -1 todo

* docs(swim-37): `lich` primitive wiring memo (memo-before-wire)

Decides shape for `captureSwim("lich", …)` before the wiring PR per
🩸's memo-before-wire standard. Maps the harness label `lich` to the
post-compaction-delegate release seam (`emitContinuationCompactionReleasedSpan`).

Resolves four design Qs:
  Q1 (timers): N/A — release helper is synchronous.
  Q2 (releasedCount): cover non-empty (production-typical) AND empty
    (defensive helper-clamp, NOT production-reachable). Pin both.
  Q3 (compaction.id): three cases — present+valid / omitted / invalid.
    Helper drops-with-log on invalid; integration-tier verifies via
    log-callback substring match.
  Q4 (drained_count): adjacent axis NOT in scope (different seam).

Proposes 6 live integration-tier tests, no separate helper-tier file
needed (integration coverage pins the helper-tier invariants through
the public surface).

Open Q deferred: cross-primitive chain-id-propagation lifecycle test
(continue_delegate post-compaction → lich release) deserves its own
primitive and memo.

Boundary discipline preserved from #405/#407/#410: STDOUT-only,
in-memory recorder, try/finally reset, no silent defaults,
snapshot-at-emit.

Refs #324
Companions:
  - docs/design/swim-37-continue-delegate-wiring-memo.md (🌻, #405)
  - swim-37 `heartbeat` memo (🌻, in flight)

* docs(swim-37): pin NaN/Infinity for compaction.id invalid path (🌊 #411 review)

Per 🌊's review on #411: extend Q3 invalid-path test list to include
`NaN` and `Infinity` alongside `-1` and `1.5`. Same defense-parity
shape as #405's `recipients` positive-integer validation.

`Number.isInteger(NaN) === false` and `Number.isInteger(Infinity) === false`,
so the existing helper guard already catches both — pinning them
observably rather than leaving the invariant implicit.

* docs(swim-37): fix test-count headers (six → eight) per 🌻 #411 nit

Header at "Proposed test list" still said six; helper-tier-vs-integration-tier
section still said "6 live tests". Both updated to eight to match the
NaN/Infinity additions from d5cf713.

…sed helper (#414)

Implements the lich primitive per the wiring memo (#411,
`docs/design/swim-37-lich-wiring-memo.md`). Maps the harness label
"lich" (\🌫️'s SOUL-file metaphor for the post-compaction
phylactery-drink) to the post-compaction-delegate release seam:
`emitContinuationCompactionReleasedSpan` at
`src/infra/continuation-tracer.ts:789`.

Changes
-------
`studies/swim-37/harness/swim-runner.ts`:
  - Import `emitContinuationCompactionReleasedSpan`
  - Add `releasedCount`, `compactionId`, `log` to `CaptureSwimOptions`
    (lich-only fields, all optional, with memo \§Q-references in JSDoc)
  - Split `heartbeat`/`lich` cases — wire `lich`, leave `heartbeat`
    as not-yet-wired pending 🌻's #412 wire PR
  - Update file header "Scope" comment to reflect new wiring state

`studies/swim-37/harness/swim-runner.test.ts`:
  - Replace 2 `it.todo` placeholders with 8 live tests in the
    "lich-shape" describe block:
      1. releasedCount=1 + compactionId=7 (production-typical)
      2. releasedCount=3 + compactionId=42 (multi-release)
      3. defensive releasedCount=0 (helper-clamp boundary, not
         production-reachable — pinned for helper-tier coverage)
      4. compactionId omitted (omission contract)
      5. compactionId=-1 (drops with log; substring match)
      6. compactionId=1.5 (non-integer drop with log)
      7. compactionId=NaN (defense parity per 🌊's #411 review)
      8. compactionId=Infinity (defense parity per 🌊's #411 review)
  - Update "refuses primitives not yet wired" test to remove `lich`
    (now wired, covered by its own describe block)

Discipline preserved
--------------------
  - STDOUT-only: no real BasicTracerProvider/OTLP machinery
  - In-memory recorder via `createInMemorySpanRecorder()`
  - try/finally tracer reset (no cross-call pollution)
  - Snapshot-at-emit: no recompute in caller
  - Drop-with-log invariant verified via log-callback substring match
  - Negative-attr assertions on omitted `compaction.id`

Validation
----------
  - swim-37 vitest project: 8 files / 138 passed / 13 todo
    (was 130 passed pre-PR; +8 new lich tests)
  - `pnpm lint` clean (0 warnings, 0 errors across all shards)

Lane discipline
---------------
  - No cross with 🌻's #412 `heartbeat` lane (different production
    helper, different file in that PR's helper-tier)
  - No reshape of #405's `continue_work`/`continue_delegate` wiring
  - Q-OPEN deferred per memo: cross-primitive lifecycle test
    (`continue_delegate` post-compaction → `lich` release sharing
    `chain.id`) deserves its own primitive shape, separate PR

Refs #324
Memo: #411 (`3d90f68b14`)

…re-wire) (#413)

* docs(swim-37): `rebase.classify` span-emission wiring memo (memo-before-wire)

Companion to PR #405 (continue_delegate memo), PR #411 (lich memo merged
3d90f68), PR #412 (heartbeat memo merged 1b84e71). Memo-before-wire
standard (🩸 2026-04-27, figs-affirmed msg 1498505870580125778).

Wires the L81 placeholder
  it.todo("CHANGELOG-byte-grep discovery channel emits drop-with-reason span")
in studies/swim-37/harness/swim-runner.test.ts onto a designed contract,
NOT just a typed signature.

Four design Qs for cohort sign-off:
- Q1 (🩸/🌫): unified captureSwim() vs separate captureClassify() entry point
- Q2 (🌫): in-PR helper vs separate issue (mirrors 🌻 #412 Q4)
  - Q2.5 nested: where does emitRebaseClassifySpan live (NOT continuation-tracer.ts)
- Q3 (🌻): 6-row it.each matrix under 🌊's split-threshold
- Q4 (cohort): does PICK-verdict ever emit (lean: yes, transparent record)

Lane discipline: NEW tracer module (rebase domain ≠ continuation domain),
NEW helper-tier sibling test file, integration-tier extends swim-runner.test.ts
§1 block. Avoids polluting continuation-tracer.ts with non-continuation helper.

Negative-assert pins on chain.id / chain.step.remaining / disabled.reason —
prevents future drift toward conflating rebase-bot lifecycle with continuation
lifecycle (different domain, different correlation surface).

* docs(swim-37): swap header commit-sha refs for memo file paths (🌻 #413 nit)

* docs(swim-37): clarify evidence.conflict.bin presence semantics (🌫 #413 nit)

Per 🌫's review nit: §2 evidence.conflict.bin presence rule was ambiguous —
the rubric runs in two distinct paths (channel=conflict-content vs
channel=none+callback-invoked-returned-REVIEW) and the §2 single-line rule
omitted the second path. Q3 matrix row 6 had it right; §2 now matches.

Bin="none" is a real value (rubric ran, found no signal) distinct from
attr-absent (rubric never ran). Pinned explicitly to prevent future readers
from collapsing the two.

---------

Co-authored-by: Ronan 🌊 <ronan@solidor.io>

…review) (#418)

Folds three flags from 🌫's #416 second-eye review (msg
`1498543295234834432`):

**Flag 1 — memo §3 validation gap:** memo specced
throw-on-bad-input matching #405/#411/#412 family-resemblance,
wire skipped it. Adds boundary throws in `captureClassify` for:
  - verdict not in REBASE_VERDICTS
  - channel not in REBASE_DISCOVERY_CHANNELS
  - pickSha < 7 hex chars (memo §3 git-prefix-min)
  - pickSha containing non-hex chars

Validation lives at the harness boundary (mirrors
`captureSwim`'s `recipients` invariant at swim-runner.ts:180);
the production helper stays drop-with-log so producer errors
don't propagate to the rebase-bot caller. Same shape as how
`captureSwim` throws on bad Options but
`emitContinuationCompactionReleasedSpan` only logs.

**Flag 2 — truncateSha ↔ memo §3 "≥7 hex chars" tension:** the
`<12 unchanged (helper does not pad)` test from #416 was
contradicting the memo's ≥7 floor. Replaced with
`passes short-but-valid (7-char) SHA through unchanged` which
honors both the memo floor AND the no-pad discipline. Memo +
code + tests now agree.

**Nit — `signal.kind: "rebase.classify"` redundant with span
name:** other helpers use the underlying primitive name as
discriminator (e.g. `"compaction-release"` for
`continuation.compaction.released`), so a downstream filter on
`signal.kind` doesn't redundantly slice the span name. Renamed
to `"rebase-classify"` (hyphen-form, matches family). Memo §2
updated with provenance note (post-#416 review).

Test additions:
  - new `describe("input validation (memo §3 throw-on-bad-input)")`
    block: 4 tests (bad verdict, bad channel, <7 char pickSha,
    non-hex pickSha)
  - existing `<12 unchanged` test deleted, replaced with valid
    7-char passes-through-unchanged

Test results: pnpm vitest run -c test/vitest/vitest.swim-37.config.ts
  → 8 files / 155 passed / 14 todo (was 151/14 from #416 at
    `526540de15`; +4 live tests, same todos)

Lane discipline: same lane as #416 follow-up (mirrors how #415
followed #414); zero overlap with 🌻's incoming heartbeat wire.

Co-authored-by: Ronan 🌊 <ronan@solidor.io>
Co-authored-by: Silas 🌫 <silas@dandelion.cult>