@hecer/yoke 1.5.1 → 1.6.1

package/TODOS.md

@@ -1,5 +1,5 @@
-# Yoke follow-up work
-
-- Add provider-native output schemas when all three CLIs expose compatible stable APIs.
-- Expand benchmark fixtures and collect multiple authenticated samples per provider/model.
-- Add signed provenance and attestations to npm and GitHub releases.
+# Yoke follow-up work
+
+- Add provider-native output schemas when all three CLIs expose compatible stable APIs.
+- Expand benchmark fixtures and collect multiple authenticated samples per provider/model.
+- Add signed provenance and attestations to npm and GitHub releases.

package/agents/docs.toml

@@ -1,6 +1,6 @@
-name = "docs"
-description = "Documentation specialist for release and API consistency."
-sandbox_mode = "workspace-write"
-developer_instructions = """
-Update only documentation required by the assigned change. Verify commands and version references against the repository.
-"""
+name = "docs"
+description = "Documentation specialist for release and API consistency."
+sandbox_mode = "workspace-write"
+developer_instructions = """
+Update only documentation required by the assigned change. Verify commands and version references against the repository.
+"""

package/agents/implementer.toml

@@ -1,6 +1,6 @@
-name = "implementer"
-description = "Implementation specialist for one scoped story."
-sandbox_mode = "workspace-write"
-developer_instructions = """
-Implement only the assigned scope. Use tests first, run verification, and do not review or commit your own work.
-"""
+name = "implementer"
+description = "Implementation specialist for one scoped story."
+sandbox_mode = "workspace-write"
+developer_instructions = """
+Implement only the assigned scope. Use tests first, run verification, and do not review or commit your own work.
+"""

package/agents/reviewer.toml

@@ -1,6 +1,6 @@
-name = "reviewer"
-description = "Read-only reviewer for correctness and acceptance criteria."
-sandbox_mode = "read-only"
-developer_instructions = """
-Review observed diffs and test evidence. Do not modify files. Return only findings grounded in evidence.
-"""
+name = "reviewer"
+description = "Read-only reviewer for correctness and acceptance criteria."
+sandbox_mode = "read-only"
+developer_instructions = """
+Review observed diffs and test evidence. Do not modify files. Return only findings grounded in evidence.
+"""

package/agents/security.toml

@@ -1,6 +1,6 @@
-name = "security"
-description = "Read-only security reviewer for changed code."
-sandbox_mode = "read-only"
-developer_instructions = """
-Inspect changed code for exploitable security regressions. Do not modify files and avoid speculative findings.
-"""
+name = "security"
+description = "Read-only security reviewer for changed code."
+sandbox_mode = "read-only"
+developer_instructions = """
+Inspect changed code for exploitable security regressions. Do not modify files and avoid speculative findings.
+"""

package/bench/README.md

@@ -1,86 +1,86 @@
-# Yoke benchmark — tokens · speed · quality
-
-Reproducible cross-runner and routing A/B benchmarks for the Yoke loop. Fixed fixture projects,
-the same PRD within each comparison, and three measured dimensions:
-
-Adaptive routing is implemented for Claude Code, Codex CLI, and Gemini CLI. The checked-in
-multi-run performance study currently measures Codex only; provider support tests are not treated
-as performance evidence for Claude or Gemini.
-
-| Dimension | How it is measured |
-|---|---|
-| **Tokens** | The loop's own provider telemetry in `.yoke/loop-status.json`, split into total input, cached input, fresh input (`input − cached`), output, and reasoning tokens when the provider reports them. Adaptive runs preserve one entry per orchestrator/worker call. Missing telemetry is `null`, never estimated. |
-| **Speed** | Wall-clock, measured by the harness from outside: total run + per-story (from `--json` NDJSON event timestamps). The loop itself stores no durations. |
-| **Quality** | Objective, not judged by any model: the fixture ships **pre-written tests** the agent never has to write (only satisfy). After the run, each story's test file is executed against the final tree. `srcLoc` (non-empty lines in `src/`) is a code-economy proxy. |
-
-## The fixture (`fixtures/string-kit`)
-
-A dependency-free ESM library with 3 stories (`slugify`, `truncate`, `titleCase`) and 16
-`node:test` assertions total. `bench-verify.mjs` is cumulative: story N runs the tests of
-stories 1…N (the loop exports `YOKE_STORY`), so later stories cannot break earlier work; the
-final quality check runs everything. No npm installs, so results measure the agent — not the
-network.
-
-## The routing fixture (`fixtures/routing-queue`)
-
-A dependency-free in-memory priority queue with two cumulative stories and 10 pre-written
-`node:test` cases covering idempotency, priority/FIFO ordering, leases, retry/backoff, dead
-letters, expired lease recovery, filters, and state statistics. It is intentionally substantial
-enough that a lower-cost worker can potentially repay one routing-controller call per story.
-
-## Running it
-
-```bash
-npm run build
-node bench/run.mjs --runner=claude   # or gemini / codex
-node bench/run.mjs --runner=codex --fixture=routing-queue --routing=off --unsafe --run-root=G:\NN-Developed\Yoke-Testground
-node bench/run.mjs --runner=codex --fixture=routing-queue --routing=on  --unsafe --run-root=G:\NN-Developed\Yoke-Testground
-node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-large-seed-2026-08-02 --routing=off --run-root=G:\NN-Developed\Yoke-Testground
-node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-large-seed-2026-08-02 --routing=on  --run-root=G:\NN-Developed\Yoke-Testground
-node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-codex-study-seed-2026-08-02 --routing=on  --label=codex-only-pair1-on  --run-root=G:\NN-Developed\Yoke-Testground
-node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-codex-study-seed-2026-08-02 --routing=off --label=codex-only-pair1-off --run-root=G:\NN-Developed\Yoke-Testground
-node bench/analyze-routing-study.mjs
-node bench/run-matrix.mjs --label=release-1.0
-node bench/output-compaction.mjs  # deterministic local gate-output benchmark; no provider call
-```
-
-Each run copies the fixture to `bench/.runs/<fixture>-<runner>-<routing>-<stamp>` (or the
-explicit `--run-root`), git-inits it,
-drives `yoke loop run --json --max=6 --timeout=10`, and writes a result JSON to
-`bench/results/`. Runs are billed against your own accounts for the agent CLIs involved.
-The matrix is sequential to avoid cross-provider load distortion. Missing CLIs and authentication
-failures are stored as honest `unavailable`/`auth-failed` rows and never presented as quality measurements.
-
-### Gate-output compaction benchmark
-
-`output-compaction.mjs` exercises only Yoke's deterministic failure-preview and artifact path.
-It emits a fixed noisy gate failure, verifies that the early compiler error and final test summary
-remain visible, and checks the stored SHA-256 digest. Its byte/token approximation is not a provider
-bill and says nothing about tool output generated inside Claude Code, Codex, or Gemini. It makes no
-comparison to Aphrodite's corpus or published ratios.
-
-`run-large.mjs` accepts an external full-repository seed, starts each arm with a separate empty
-routing registry, junctions this checkout's dependencies, and replays the seed's original
-acceptance tests under fresh filenames after the agent run. This prevents an agent editing a
-visible test from turning into false benchmark evidence. A seed can provide `bench-acceptance.json`
-to declare its fixture identity and hidden-test files. `analyze-routing-study.mjs` validates and
-aggregates the checked-in three-pair Codex-only study.
-
-## Caveats (read before quoting numbers)
-
-- Agent runs are stochastic. Older rows are N=1; the 2026-08-02 Codex-only study uses three
-  alternating-order pairs per arm. Re-run before setting broad policy defaults.
-- Compare routing on/off only when fixture, parent model/effort, permissions, native-subagent
-  policy, and host load are controlled. The checked-in routing fixture uses `runner.bare: true`
-  to exclude personal MCP/plugin startup from both sides.
-- Model identity matters more than CLI identity: `tokens.model` records what actually served
-  the run. Different default models per CLI make "claude vs gemini" really "model X vs model Y".
-- Codex input telemetry includes cache reads. Compare both total input and fresh input; do not
-  treat cached input as equivalent to newly processed input. Dollar cost is reported only when
-  the provider emits it—Yoke does not guess prices from a model name.
-- The fixture is deliberately small (a loop-overhead + basic-competence probe, minutes not
-  hours). It does not measure large-context refactoring, UI work, or long-horizon planning.
-- Cumulative verify means a story's duration includes fixing any regressions it caused.
-
-Results live in [`results/`](results/) — one JSON per run, summarized in
-[`RESULTS.md`](RESULTS.md).
+# Yoke benchmark — tokens · speed · quality
+
+Reproducible cross-runner and routing A/B benchmarks for the Yoke loop. Fixed fixture projects,
+the same PRD within each comparison, and three measured dimensions:
+
+Adaptive routing is implemented for Claude Code, Codex CLI, and Gemini CLI. The checked-in
+multi-run performance study currently measures Codex only; provider support tests are not treated
+as performance evidence for Claude or Gemini.
+
+| Dimension | How it is measured |
+|---|---|
+| **Tokens** | The loop's own provider telemetry in `.yoke/loop-status.json`, split into total input, cached input, fresh input (`input − cached`), output, and reasoning tokens when the provider reports them. Adaptive runs preserve one entry per orchestrator/worker call. Missing telemetry is `null`, never estimated. |
+| **Speed** | Wall-clock, measured by the harness from outside: total run + per-story (from `--json` NDJSON event timestamps). The loop itself stores no durations. |
+| **Quality** | Objective, not judged by any model: the fixture ships **pre-written tests** the agent never has to write (only satisfy). After the run, each story's test file is executed against the final tree. `srcLoc` (non-empty lines in `src/`) is a code-economy proxy. |
+
+## The fixture (`fixtures/string-kit`)
+
+A dependency-free ESM library with 3 stories (`slugify`, `truncate`, `titleCase`) and 16
+`node:test` assertions total. `bench-verify.mjs` is cumulative: story N runs the tests of
+stories 1…N (the loop exports `YOKE_STORY`), so later stories cannot break earlier work; the
+final quality check runs everything. No npm installs, so results measure the agent — not the
+network.
+
+## The routing fixture (`fixtures/routing-queue`)
+
+A dependency-free in-memory priority queue with two cumulative stories and 10 pre-written
+`node:test` cases covering idempotency, priority/FIFO ordering, leases, retry/backoff, dead
+letters, expired lease recovery, filters, and state statistics. It is intentionally substantial
+enough that a lower-cost worker can potentially repay one routing-controller call per story.
+
+## Running it
+
+```bash
+npm run build
+node bench/run.mjs --runner=claude   # or gemini / codex
+node bench/run.mjs --runner=codex --fixture=routing-queue --routing=off --unsafe --run-root=G:\NN-Developed\Yoke-Testground
+node bench/run.mjs --runner=codex --fixture=routing-queue --routing=on  --unsafe --run-root=G:\NN-Developed\Yoke-Testground
+node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-large-seed-2026-08-02 --routing=off --run-root=G:\NN-Developed\Yoke-Testground
+node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-large-seed-2026-08-02 --routing=on  --run-root=G:\NN-Developed\Yoke-Testground
+node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-codex-study-seed-2026-08-02 --routing=on  --label=codex-only-pair1-on  --run-root=G:\NN-Developed\Yoke-Testground
+node bench/run-large.mjs --seed=G:\NN-Developed\Yoke-Testground\yoke-codex-study-seed-2026-08-02 --routing=off --label=codex-only-pair1-off --run-root=G:\NN-Developed\Yoke-Testground
+node bench/analyze-routing-study.mjs
+node bench/run-matrix.mjs --label=release-1.0
+node bench/output-compaction.mjs  # deterministic local gate-output benchmark; no provider call
+```
+
+Each run copies the fixture to `bench/.runs/<fixture>-<runner>-<routing>-<stamp>` (or the
+explicit `--run-root`), git-inits it,
+drives `yoke loop run --json --max=6 --timeout=10`, and writes a result JSON to
+`bench/results/`. Runs are billed against your own accounts for the agent CLIs involved.
+The matrix is sequential to avoid cross-provider load distortion. Missing CLIs and authentication
+failures are stored as honest `unavailable`/`auth-failed` rows and never presented as quality measurements.
+
+### Gate-output compaction benchmark
+
+`output-compaction.mjs` exercises only Yoke's deterministic failure-preview and artifact path.
+It emits a fixed noisy gate failure, verifies that the early compiler error and final test summary
+remain visible, and checks the stored SHA-256 digest. Its byte/token approximation is not a provider
+bill and says nothing about tool output generated inside Claude Code, Codex, or Gemini. It makes no
+comparison to Aphrodite's corpus or published ratios.
+
+`run-large.mjs` accepts an external full-repository seed, starts each arm with a separate empty
+routing registry, junctions this checkout's dependencies, and replays the seed's original
+acceptance tests under fresh filenames after the agent run. This prevents an agent editing a
+visible test from turning into false benchmark evidence. A seed can provide `bench-acceptance.json`
+to declare its fixture identity and hidden-test files. `analyze-routing-study.mjs` validates and
+aggregates the checked-in three-pair Codex-only study.
+
+## Caveats (read before quoting numbers)
+
+- Agent runs are stochastic. Older rows are N=1; the 2026-08-02 Codex-only study uses three
+  alternating-order pairs per arm. Re-run before setting broad policy defaults.
+- Compare routing on/off only when fixture, parent model/effort, permissions, native-subagent
+  policy, and host load are controlled. The checked-in routing fixture uses `runner.bare: true`
+  to exclude personal MCP/plugin startup from both sides.
+- Model identity matters more than CLI identity: `tokens.model` records what actually served
+  the run. Different default models per CLI make "claude vs gemini" really "model X vs model Y".
+- Codex input telemetry includes cache reads. Compare both total input and fresh input; do not
+  treat cached input as equivalent to newly processed input. Dollar cost is reported only when
+  the provider emits it—Yoke does not guess prices from a model name.
+- The fixture is deliberately small (a loop-overhead + basic-competence probe, minutes not
+  hours). It does not measure large-context refactoring, UI work, or long-horizon planning.
+- Cumulative verify means a story's duration includes fixing any regressions it caused.
+
+Results live in [`results/`](results/) — one JSON per run, summarized in
+[`RESULTS.md`](RESULTS.md).

package/bench/RESULTS.md

@@ -1,22 +1,22 @@
-# Benchmark results
-
-Result schema v1 records fixture version, sample label, permission profile, telemetry/model
-availability, verdict/blocker, conflicts, wall time, iterations, and final fixture-test status.
-
-Fixture `string-kit` (3 stories, 16 pre-written assertions). Methodology and caveats:
-[README.md](README.md). One row per run — raw JSON in [`results/`](results/).
-
-## Runs
-
-| Date | Runner | Model (reported) | Result | Wall-clock | Input tok | Output tok | First-pass stories | src LOC |
-|---|---|---|---|---|---|---|---|---|
-| 2026-07-27 | claude | unavailable | ⛔ blocked before implementation; zero synthetic usage is not counted | 11 s | — | — | 0/3 | 3 |
-| 2026-07-27 | codex | unavailable | ⛔ CLI probe failed on Windows with access denied | — | — | — | — | — |
-| 2026-07-27 | gemini | unavailable | ⛔ runner exited before implementation | 14 s | — | — | 0/3 | 3 |
-| 2026-07-10 | claude | claude-opus-4-8 | ✅ 3/3 complete | 4 m 28 s | 42 957 | 9 817 | 3/3 (1 iteration each) | 39 |
-| 2026-07-10 | gemini | — | ⛔ blocked: CLI not authenticated on the bench machine (headless needs `GEMINI_API_KEY` or configured OAuth) | — | — | — | — | — |
-| — | codex | — | ⛔ not installed on the bench machine | — | — | — | — | — |
-
+# Benchmark results
+
+Result schema v1 records fixture version, sample label, permission profile, telemetry/model
+availability, verdict/blocker, conflicts, wall time, iterations, and final fixture-test status.
+
+Fixture `string-kit` (3 stories, 16 pre-written assertions). Methodology and caveats:
+[README.md](README.md). One row per run — raw JSON in [`results/`](results/).
+
+## Runs
+
+| Date | Runner | Model (reported) | Result | Wall-clock | Input tok | Output tok | First-pass stories | src LOC |
+|---|---|---|---|---|---|---|---|---|
+| 2026-07-27 | claude | unavailable | ⛔ blocked before implementation; zero synthetic usage is not counted | 11 s | — | — | 0/3 | 3 |
+| 2026-07-27 | codex | unavailable | ⛔ CLI probe failed on Windows with access denied | — | — | — | — | — |
+| 2026-07-27 | gemini | unavailable | ⛔ runner exited before implementation | 14 s | — | — | 0/3 | 3 |
+| 2026-07-10 | claude | claude-opus-4-8 | ✅ 3/3 complete | 4 m 28 s | 42 957 | 9 817 | 3/3 (1 iteration each) | 39 |
+| 2026-07-10 | gemini | — | ⛔ blocked: CLI not authenticated on the bench machine (headless needs `GEMINI_API_KEY` or configured OAuth) | — | — | — | — | — |
+| — | codex | — | ⛔ not installed on the bench machine | — | — | — | — | — |
+
 Per-story wall-clock (claude run): STORY-1 72 s · STORY-2 114 s · STORY-3 81 s. Every story
 passed verify on the first iteration; the final quality check (all 16 assertions on the final
 tree, outside the loop) is green.
@@ -115,22 +115,22 @@ zero-token deterministic SELF fast path for clearly high-risk architecture/priva
 This is again N=1. Raw valid rows:
 [`routing off`](results/yoke-large-codex-routing-off-2026-08-01T22-58-01.json) and
 [`routing on`](results/yoke-large-codex-routing-on-2026-08-01T23-09-39.json).
-
-The 2026-07-27 release matrix produced no quality measurement: Claude and Gemini exited
-before changing the fixture, and the Codex executable could not be probed in this Windows
-environment. Raw rows preserve the exact runner diagnostics; no scores were inferred.
-
-## What the harness caught before producing a single number
-
-Building an honest benchmark is itself a verification pass. The first runs found two real
-Yoke bugs, both fixed in 0.3.0:
-
-1. **Availability-probe timeout** — `gemini --version` cold-starts in ~5.8 s on Windows; the
-   5 s probe timeout misreported an installed CLI as "not found on PATH". Now 20 s.
-2. **Gemini invocation** — Gemini CLI 0.33+ requires a value after `-p`; the runner's bare
-   `-p --yolo` died with "Not enough arguments following: p". The runner now relies on piped
-   stdin (which selects headless mode) and passes only `--yolo`.
-
+
+The 2026-07-27 release matrix produced no quality measurement: Claude and Gemini exited
+before changing the fixture, and the Codex executable could not be probed in this Windows
+environment. Raw rows preserve the exact runner diagnostics; no scores were inferred.
+
+## What the harness caught before producing a single number
+
+Building an honest benchmark is itself a verification pass. The first runs found two real
+Yoke bugs, both fixed in 0.3.0:
+
+1. **Availability-probe timeout** — `gemini --version` cold-starts in ~5.8 s on Windows; the
+   5 s probe timeout misreported an installed CLI as "not found on PATH". Now 20 s.
+2. **Gemini invocation** — Gemini CLI 0.33+ requires a value after `-p`; the runner's bare
+   `-p --yolo` died with "Not enough arguments following: p". The runner now relies on piped
+   stdin (which selects headless mode) and passes only `--yolo`.
+
 ## Reading the numbers
 
 - Tokens come from provider telemetry when available; missing telemetry is recorded as missing,

package/bench/output-compaction.mjs

@@ -1,65 +1,65 @@
-import { createHash } from 'node:crypto'
-import { existsSync, mkdtempSync, readFileSync, rmSync } from 'node:fs'
-import { tmpdir } from 'node:os'
-import { join, resolve } from 'node:path'
-import { fileURLToPath } from 'node:url'
-
-async function runtimeModules() {
-  const builtCompact = new URL('../dist/output/compact.js', import.meta.url)
-  const builtArtifact = new URL('../dist/output/artifact.js', import.meta.url)
-  const useBuild = existsSync(fileURLToPath(builtCompact)) && existsSync(fileURLToPath(builtArtifact))
-  return Promise.all([
-    import(useBuild ? builtCompact.href : new URL('../src/output/compact.ts', import.meta.url).href),
-    import(useBuild ? builtArtifact.href : new URL('../src/output/artifact.ts', import.meta.url).href),
-  ])
-}
-
-function fixture() {
-  return [
-    '=== stdout ===',
-    'compiling application',
-    'src/core.ts:17: error TS2304: Cannot find name ImportantWidget',
-    'the failing call is part of the checkout flow',
-    ...Array.from({ length: 600 }, (_, index) => `progress shard=${index % 12} status=unchanged cache=hit`),
-    '=== stderr ===',
-    'Tests: 1 failed, 249 passed, 250 total',
-  ].join('\n')
-}
-
-export async function runOutputCompactionBenchmark() {
-  const [{ compactCommandOutput }, { writeOutputArtifact }] = await runtimeModules()
-  const raw = fixture()
-  const previewBudgetBytes = 512
-  const compacted = compactCommandOutput(raw, { previewBytes: previewBudgetBytes })
-  const dir = mkdtempSync(join(tmpdir(), 'yoke-output-bench-'))
-  try {
-    const artifact = writeOutputArtifact(dir, raw, { phase: 'verify', storyId: 'BENCH' })
-    const stored = readFileSync(join(dir, ...artifact.relativePath.split('/')))
-    const storedDigest = createHash('sha256').update(stored).digest('hex')
-    const previewBytes = Buffer.byteLength(compacted.preview)
-    const referencedBytes = previewBytes + Buffer.byteLength(artifact.marker)
-    return {
-      fixture: 'gate-output-v1',
-      rawBytes: Buffer.byteLength(raw),
-      rawApproxTokens: Math.ceil(Buffer.byteLength(raw) / 4),
-      previewBudgetBytes,
-      previewBytes,
-      previewApproxTokens: Math.ceil(previewBytes / 4),
-      referencedBytes,
-      compressionRatio: Number((Buffer.byteLength(raw) / referencedBytes).toFixed(2)),
-      earlyErrorRetained: compacted.preview.includes('error TS2304'),
-      finalSummaryRetained: compacted.preview.includes('Tests: 1 failed, 249 passed'),
-      digestRoundTrip: storedDigest === artifact.sha256,
-    }
-  } finally {
-    rmSync(dir, { recursive: true, force: true })
-  }
-}
-
-if (process.argv[1] && resolve(process.argv[1]) === fileURLToPath(import.meta.url)) {
-  const result = await runOutputCompactionBenchmark()
-  console.log(JSON.stringify(result, null, 2))
-  if (!result.earlyErrorRetained || !result.finalSummaryRetained || !result.digestRoundTrip || result.previewBytes > result.previewBudgetBytes) {
-    process.exitCode = 1
-  }
-}
+import { createHash } from 'node:crypto'
+import { existsSync, mkdtempSync, readFileSync, rmSync } from 'node:fs'
+import { tmpdir } from 'node:os'
+import { join, resolve } from 'node:path'
+import { fileURLToPath } from 'node:url'
+
+async function runtimeModules() {
+  const builtCompact = new URL('../dist/output/compact.js', import.meta.url)
+  const builtArtifact = new URL('../dist/output/artifact.js', import.meta.url)
+  const useBuild = existsSync(fileURLToPath(builtCompact)) && existsSync(fileURLToPath(builtArtifact))
+  return Promise.all([
+    import(useBuild ? builtCompact.href : new URL('../src/output/compact.ts', import.meta.url).href),
+    import(useBuild ? builtArtifact.href : new URL('../src/output/artifact.ts', import.meta.url).href),
+  ])
+}
+
+function fixture() {
+  return [
+    '=== stdout ===',
+    'compiling application',
+    'src/core.ts:17: error TS2304: Cannot find name ImportantWidget',
+    'the failing call is part of the checkout flow',
+    ...Array.from({ length: 600 }, (_, index) => `progress shard=${index % 12} status=unchanged cache=hit`),
+    '=== stderr ===',
+    'Tests: 1 failed, 249 passed, 250 total',
+  ].join('\n')
+}
+
+export async function runOutputCompactionBenchmark() {
+  const [{ compactCommandOutput }, { writeOutputArtifact }] = await runtimeModules()
+  const raw = fixture()
+  const previewBudgetBytes = 512
+  const compacted = compactCommandOutput(raw, { previewBytes: previewBudgetBytes })
+  const dir = mkdtempSync(join(tmpdir(), 'yoke-output-bench-'))
+  try {
+    const artifact = writeOutputArtifact(dir, raw, { phase: 'verify', storyId: 'BENCH' })
+    const stored = readFileSync(join(dir, ...artifact.relativePath.split('/')))
+    const storedDigest = createHash('sha256').update(stored).digest('hex')
+    const previewBytes = Buffer.byteLength(compacted.preview)
+    const referencedBytes = previewBytes + Buffer.byteLength(artifact.marker)
+    return {
+      fixture: 'gate-output-v1',
+      rawBytes: Buffer.byteLength(raw),
+      rawApproxTokens: Math.ceil(Buffer.byteLength(raw) / 4),
+      previewBudgetBytes,
+      previewBytes,
+      previewApproxTokens: Math.ceil(previewBytes / 4),
+      referencedBytes,
+      compressionRatio: Number((Buffer.byteLength(raw) / referencedBytes).toFixed(2)),
+      earlyErrorRetained: compacted.preview.includes('error TS2304'),
+      finalSummaryRetained: compacted.preview.includes('Tests: 1 failed, 249 passed'),
+      digestRoundTrip: storedDigest === artifact.sha256,
+    }
+  } finally {
+    rmSync(dir, { recursive: true, force: true })
+  }
+}
+
+if (process.argv[1] && resolve(process.argv[1]) === fileURLToPath(import.meta.url)) {
+  const result = await runOutputCompactionBenchmark()
+  console.log(JSON.stringify(result, null, 2))
+  if (!result.earlyErrorRetained || !result.finalSummaryRetained || !result.digestRoundTrip || result.previewBytes > result.previewBudgetBytes) {
+    process.exitCode = 1
+  }
+}

package/bench/result-schema.mjs

@@ -1,12 +1,12 @@
-export const requiredResultFields = [
-  'schemaVersion', 'fixtureVersion', 'runner', 'sampleLabel', 'permissionProfile',
-  'usageAvailable', 'modelAvailable', 'verdict', 'conflicts', 'wallClockMs',
-  'iterations', 'finalTestsPass',
-]
-
-export function validateResult(result) {
-  const missing = requiredResultFields.filter(key => !(key in result))
-  if (missing.length) throw new Error(`benchmark result missing: ${missing.join(', ')}`)
-  if (!['completed', 'blocked', 'unavailable', 'auth-failed'].includes(result.verdict)) throw new Error(`invalid verdict: ${result.verdict}`)
-  return result
-}
+export const requiredResultFields = [
+  'schemaVersion', 'fixtureVersion', 'runner', 'sampleLabel', 'permissionProfile',
+  'usageAvailable', 'modelAvailable', 'verdict', 'conflicts', 'wallClockMs',
+  'iterations', 'finalTestsPass',
+]
+
+export function validateResult(result) {
+  const missing = requiredResultFields.filter(key => !(key in result))
+  if (missing.length) throw new Error(`benchmark result missing: ${missing.join(', ')}`)
+  if (!['completed', 'blocked', 'unavailable', 'auth-failed'].includes(result.verdict)) throw new Error(`invalid verdict: ${result.verdict}`)
+  return result
+}

package/bench/results/claude-2026-07-27T18-03-26.json

@@ -1,50 +1,50 @@
-{
-  "schemaVersion": 1,
-  "fixtureVersion": "string-kit@1",
-  "runner": "claude",
-  "sampleLabel": "matrix-2026-07-27T18:03:26.202Z",
-  "permissionProfile": "safe",
-  "yokeVersion": "1.0.0",
-  "fixture": "string-kit",
-  "startedAt": "2026-07-27T18:03:26.598Z",
-  "wallClockMs": 11345,
-  "exitCode": 1,
-  "finalState": "blocked",
-  "verdict": "blocked",
-  "blocker": "Claude runner exited before implementation; fixture verification remained red.",
-  "conflicts": 0,
-  "iterations": 1,
-  "finalTestsPass": false,
-  "progress": {
-    "passed": 0,
-    "total": 3
-  },
-  "usageAvailable": false,
-  "modelAvailable": false,
-  "tokens": {
-    "inputTokens": 0,
-    "outputTokens": 0,
-    "model": "<synthetic>"
-  },
-  "stories": [
-    {
-      "id": "STORY-1",
-      "durationMs": 10954,
-      "iterations": 1,
-      "finalTestsPass": false
-    },
-    {
-      "id": "STORY-2",
-      "durationMs": null,
-      "iterations": 0,
-      "finalTestsPass": false
-    },
-    {
-      "id": "STORY-3",
-      "durationMs": null,
-      "iterations": 0,
-      "finalTestsPass": false
-    }
-  ],
-  "srcLoc": 3
-}
+{
+  "schemaVersion": 1,
+  "fixtureVersion": "string-kit@1",
+  "runner": "claude",
+  "sampleLabel": "matrix-2026-07-27T18:03:26.202Z",
+  "permissionProfile": "safe",
+  "yokeVersion": "1.0.0",
+  "fixture": "string-kit",
+  "startedAt": "2026-07-27T18:03:26.598Z",
+  "wallClockMs": 11345,
+  "exitCode": 1,
+  "finalState": "blocked",
+  "verdict": "blocked",
+  "blocker": "Claude runner exited before implementation; fixture verification remained red.",
+  "conflicts": 0,
+  "iterations": 1,
+  "finalTestsPass": false,
+  "progress": {
+    "passed": 0,
+    "total": 3
+  },
+  "usageAvailable": false,
+  "modelAvailable": false,
+  "tokens": {
+    "inputTokens": 0,
+    "outputTokens": 0,
+    "model": "<synthetic>"
+  },
+  "stories": [
+    {
+      "id": "STORY-1",
+      "durationMs": 10954,
+      "iterations": 1,
+      "finalTestsPass": false
+    },
+    {
+      "id": "STORY-2",
+      "durationMs": null,
+      "iterations": 0,
+      "finalTestsPass": false
+    },
+    {
+      "id": "STORY-3",
+      "durationMs": null,
+      "iterations": 0,
+      "finalTestsPass": false
+    }
+  ],
+  "srcLoc": 3
+}

package/bench/results/codex-unavailable-1785175418318.json

@@ -1,15 +1,15 @@
-{
-  "schemaVersion": 1,
-  "fixtureVersion": "string-kit@1",
-  "runner": "codex",
-  "sampleLabel": "matrix-2026-07-27T18:03:26.202Z",
-  "permissionProfile": "safe",
-  "usageAvailable": false,
-  "modelAvailable": false,
-  "verdict": "unavailable",
-  "blocker": "Zugriff verweigert",
-  "conflicts": 0,
-  "wallClockMs": null,
-  "iterations": 0,
-  "finalTestsPass": false
-}
+{
+  "schemaVersion": 1,
+  "fixtureVersion": "string-kit@1",
+  "runner": "codex",
+  "sampleLabel": "matrix-2026-07-27T18:03:26.202Z",
+  "permissionProfile": "safe",
+  "usageAvailable": false,
+  "modelAvailable": false,
+  "verdict": "unavailable",
+  "blocker": "Zugriff verweigert",
+  "conflicts": 0,
+  "wallClockMs": null,
+  "iterations": 0,
+  "finalTestsPass": false
+}