@surlymochan/evalid 0.5.8 → 0.5.10

package/CHANGELOG.md

@@ -5,6 +5,22 @@ All notable changes to this project are documented here. Detailed notes live in
 
 ## [Unreleased]
 
+## [0.5.10] - 2026-11-08
+
+### Added
+- **Intake atoms coverage (#60)**: `src/core/atoms/intake/preview-or-refine.js` (G1 `intake.preview_or_refine`) and `src/core/atoms/intake/recommend.js` (G5 `intake.recommend`) were at 0% coverage. Added `test/intake-atoms.test.js` (8 tests) covering intake preview/refine file writing with mode/summary/answers/constraints, context title fallback for the summary, change_id from input, the missing-change_id error, workflow recommendation (yolo for docs with no coverage gate, corps with a human checkpoint for payments) plus `workflow_recommendation.json` / `policy_effective.json` writing, and registry wiring for both atoms. Test evidence: new file 8/8 pass; full suite green on the merged tree; coverage after: preview-or-refine 100% (19/19), recommend 100% (20/20) (both were 0% before); all-files statements 86.78% → 86.97%.
+- **Phase-run fail atom coverage (#61)**: `src/core/atoms/phase-run/fail.js` (D3d `phase_run.fail`, ~40 lines of failure-path state machine) was at 0% coverage. Added 7 tests to `test/heavy-atoms.test.js` alongside the existing D3 lifecycle tests, covering the success path (explicit session_id + summary + detail, session marked failed with stop_reason and failure persisted), resolving the most recent running session without a session_id, session_id from `context.runtime.phaseRunSessionId`, stop_reason fallback and the default `phase_run_failed` summary, an explicit `data_file` custom store, and the two error paths (missing mission, no session for the mission). Test evidence: RED→GREEN (stubbed-throwing implementation failed all 6 initial tests, then green after restore); `node test/heavy-atoms.test.js` 54 pass / 0 fail; coverage after: fail.js 100% statements / 94.44% branches / 100% functions / 100% lines (was 0% on all four; the single uncovered branch is the `process.cwd()` fallback, untested by design to avoid polluting the repo state).
+
+Acceptance debt, stated plainly: the bench-number acceptance for P1-2 / P1-3 / P2-1 / P2-3 still awaits the MiniMax full-matrix batch (15 instances × 4 arms × reps 3), which is stopped on MiniMax Token Plan quota exhaustion. No bench numbers are claimed in this release.
+
+## [0.5.9] - 2026-11-08
+
+### Added
+- **Design atoms coverage (#57)**: `src/agent_team/atoms/design/` H2b/H2d/H2e atoms (competitor reconstruction review, reference surface lock, reconstruction pack) had no tests. Added `test/design-atoms.test.js` (10 tests) covering contract freezing, primary-surface selection, non-array input handling, lock-file writing and pack building. Test evidence: full suite green on the merged tree.
+- **Memory atoms coverage (#58)**: `src/core/memory-state.js` and `src/core/atoms/memory/manage.js` / `mount.js` were at 0% coverage. Added `test/memory-atoms.test.js` (21 tests) covering create/list/mount/resolve defaults and filters, mission resolution via input, runtime and bindings file, and manage/mount against a temp-dir sqlite store (the `process.cwd()` fallback is exercised via `AS_XFLOW_DATA_FILE` pointed at a temp path, so no workspace pollution). No implementation changes: two initially failing expectations were test-side misreadings of `ensureArray` (truthy non-array values are wrapped, not reset). Test evidence: `npm run coverage` 404 tests / 403 pass / 0 fail / 1 skipped (pre-existing); coverage after: memory-state 100/100/100/100, manage 100/100/100/100, mount 100/100/100/100 (all three were 0% before); all-files statements 86.28% → 86.78%.
+
+Acceptance debt, stated plainly: the bench-number acceptance for P1-2 / P1-3 / P2-1 / P2-3 still awaits the MiniMax full-matrix batch (15 instances × 4 arms × reps 3), which is stopped on MiniMax Token Plan quota exhaustion. No bench numbers are claimed in this release.
+
 ## [0.5.8] - 2026-11-08
 
 ### Added

package/RELEASE_NOTES.md

@@ -1,5 +1,25 @@
 # Release Notes
 
+## 0.5.10 - Intake and phase-run fail atoms test coverage
+
+Patch release. Two test-coverage additions since 0.5.9, both verified by the full test suite. No runtime behaviour changed.
+
+**Intake atoms coverage (#60).** G1 `intake.preview_or_refine` and G5 `intake.recommend` now have 8 tests in `test/intake-atoms.test.js`, covering intake preview/refine file writing, summary title fallback, change_id handling, workflow recommendation (yolo for docs, corps with a human checkpoint for payments) and registry wiring. Coverage went from 0% to 100% on both files (19/19 and 20/20); all-files statements 86.78% → 86.97%. Test evidence: new file 8/8 pass; full suite green on the merged tree.
+
+**Phase-run fail atom coverage (#61).** D3d `phase_run.fail` (`src/core/atoms/phase-run/fail.js`) now has 7 tests in `test/heavy-atoms.test.js`, covering the success path, session resolution (explicit id, runtime id, most recent running session), stop_reason fallback, custom store via `data_file`, and both error paths. Coverage went from 0% to 100% statements / 94.44% branches / 100% functions / 100% lines. Test evidence: RED→GREEN verified; `node test/heavy-atoms.test.js` 54 pass / 0 fail.
+
+Acceptance debt, stated plainly: the bench-number acceptance for P1-2, P1-3, P2-1 and P2-3 — the wall-clock/token/resolved lines from issue #5 — still awaits the MiniMax full-matrix batch (15 instances × 4 arms × reps 3), which is stopped on MiniMax Token Plan quota exhaustion. No bench numbers are claimed in this release.
+
+## 0.5.9 - Design and memory atoms test coverage
+
+Patch release. Two test-coverage additions since 0.5.8, both verified by the full test suite. No runtime behaviour changed.
+
+**Design atoms coverage (#57).** H2b/H2d/H2e design atoms (competitor reconstruction review, reference surface lock, reconstruction pack) now have 10 tests in `test/design-atoms.test.js`, covering contract freezing, primary-surface selection, non-array input handling, lock-file writing and pack building.
+
+**Memory atoms coverage (#58).** `src/core/memory-state.js` and the `memory/manage` and `memory/mount` atoms now have 21 tests in `test/memory-atoms.test.js`, covering create/list/mount/resolve defaults and filters, mission resolution via input, runtime and bindings file, and manage/mount against a temp-dir sqlite store. Coverage went from 0% to 100/100/100/100 on all three files. Test evidence: `npm run coverage` 404 tests / 403 pass / 0 fail / 1 skipped (pre-existing).
+
+Acceptance debt, stated plainly: the bench-number acceptance for P1-2, P1-3, P2-1 and P2-3 — the wall-clock/token/resolved lines from issue #5 — still awaits the MiniMax full-matrix batch (15 instances × 4 arms × reps 3), which is stopped on MiniMax Token Plan quota exhaustion. No bench numbers are claimed in this release.
+
 ## 0.5.8 - README command reference; CI actions to v7
 
 Patch release. One docs addition and one CI maintenance bump since 0.5.7, both verified by the full test suite.

package/docs/launch-dossier.md

@@ -1,6 +1,6 @@
 # evalid Launch Dossier
 
-Generated: 2026-11-07T21:23:51.607Z
+Generated: 2026-11-08T01:34:24.683Z
 Mode: post_publish
 Verdict: conditionally ready: engine and adoption evidence are strong, registry publication remains required for a splash launch
 Launch ready: not yet
@@ -117,7 +117,7 @@ Evidence:
 - Adoption records: 1; status: pass
 - Goal audit: pass; goal: Keep evalid credible as an open-source agent delivery runtime: repo-owned workflow contracts, executable verification gates, durable project context, and honest benchmark evidence.
 - Package evidence: missing; source: registry_json
-- Package name/version: @surlymochan/evalid@0.5.8
+- Package name/version: @surlymochan/evalid@0.5.10
 
 ## Passing Evidence
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@surlymochan/evalid",
-  "version": "0.5.8",
+  "version": "0.5.10",
   "description": "Claimed != verified: a runtime verification layer for coding agents. It seals each test run and replays it from a clean snapshot, with a zero-judge gate pipeline (spec clarification, code review, TDD proof) and the xflow CLI.",
   "keywords": [
     "ai-agents",