Files
pipdock/docs/ROADMAP.md
T
poli0981andClaude Opus 5 4720e1ccb5 docs: close Phase 3, with the numbers it was measured at
The exit criterion was *"CODE-HEALTH flows pass on two real projects (one pyproject, one
requirements-only)"*. Both are recorded in the stage table with times from a release build
rather than ticked: a requirements-only project reports 1 dependency, 3 dead-code and 3 lint
findings in 1.4 s; one declaring nothing reports `ran: [vulture, ruff]` with an empty
`problems` in 1.3 s, because deptry cannot answer its question there and is correctly not
run. The fix path's three refusals were exercised on a seeded repository.

UI-SPEC §8 gains the two rules P6 established and the reason each exists: `←`/`→` are the
ARIA grid pattern's roving tabindex, added because making the row the only tab stop had left
*Pin* and *Remove* reachable by mouse only; and `Enter` on a row is the non-destructive
primary, because a row has one reversible annotation and one irreversible removal.

CLI-SPEC §6's amendment said *"`health` is still the M3 stub"* — stale since P3, and now
carrying the `--fix` output contract instead.

CLAUDE.md records the tab-stop rule beside the other things this project learned by running,
and the status line moves to Phase 4.

**Still owed, and it needs the real Tauri runtime rather than the stubbed bridge:** the OS
folder picker, a ruff rule link (to confirm the `docs.astral.sh` allowlist entry actually
works — it fails silently if not), and the fix dialog end to end. All three are named in the
ROADMAP so they are not rediscovered.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-13 19:48:48 +07:00

60 KiB
Raw Blame History

PipDock — Roadmap

Version 0.1 · 2026-07-17

Phase 0 — Spike week (before any feature code) — RUN 2026-07-27

The entire architecture stands on assumptions that must be verified on a real machine first (the FrameLedger/QuoteAtlas discipline). Each spike is a small script + captured fixtures committed to crates/pipdock-core/tests/fixtures/.

Outcomes — full write-up in spikes/README.md.

Spike Verdict The one thing that changed
SP-1 GO — ship both engines Neither engine holds packages back on install -U <pkg>; both silently break installed dependents. Every plan must restate the full installed set as explicit requirements. uv's conflict attribution is better than pip's; uv writes its plan to stderr, not stdout. uv minimum pinned at 0.10.0.
SP-2 Corpus captured; blocker found and fixed pip install --dry-run --report - crashes with UnicodeEncodeError on Windows/cp1252 (pip 25 and 26). Every pip invocation must set PYTHONIOENCODING=utf-8 + PYTHONUTF8=1. Also: yanked releases exit 0 (a preview warning, not a failure), and pip/uv disagree on Requires-Python enforcement.
SP-3 PASS 858 k projects; cold refresh 2.95 s (budget 60 s); worst keystroke 42.1 ms (budget 50 ms) — but only in memory: scanning SQLite per keystroke costs 218 ms. Margin is thin on fast hardware, and raw nucleo ranking puts requests-ntlm above requests.
SP-4 Answered --path cannot be combined with --no-deps; freeze-file mode (-r <freeze> --no-deps -f json) is the only option. SECURITY §6 confirmed. Findings arrive with duplicate advisory ids.
SP-5 Answered by SP-1 No engine reports held-back items at all. Resolved version, latest version and blocker must all be derived by PipDock — for pip as much as for uv, contrary to ARCHITECTURE §3's implication.
SP-6 Answered uv python list returns downloadable entries and duplicate shims; env_hash must lowercase the interpreter path on Windows or pins and snapshots split in two. probe.py -I hides 24 user-site packages on a system Python — open decision. Microsoft Store Python still unverified.

Both adapter-behaviour questions were decided 2026-07-27 and are implemented: PipDock enforces Requires-Python itself so the preview is engine-independent (crates/pipdock-core/src/compat.rs, ARCHITECTURE §3), and probe.py keeps -I while reporting hidden_user_site so the Installed screen can disclose the gap accurately (SECURITY §2). Two non-blocking questions remain at the end of spikes/README.md: search ranking scope, and how to obtain the five error fixtures this machine cannot produce.

# Spike Question to answer Exit criterion
SP-1 uv dry-run shape Exact stdout of uv pip install -U --dry-run (current + previous uv) across: clean upgrade, held-back, impossible. Is held-back attribution derivable, or does uv only show the resolved set? Fixture set committed; adapter parsing strategy written; uv minimum version pinned. Go/no-go for launching with both engines — if uv's plan output is too lossy, v1.0 ships pip-primary with uv behind a "beta engine" flag.
SP-2 pip report & stderr corpus Capture --dry-run --report JSON + real stderr for every ERROR-CATALOG build/network/permission case (broken sdist, MSVC-missing, PEP 668, locked file, yanked, hash mismatch). ≥ 1 fixture per launch catalog code.
SP-3 PEP 691 index economics Full name-index size, download time on VN broadband, SQLite ingest time, fuzzy-search latency on ~600 k names. < 60 s cold refresh, < 50 ms search on the i7-14700KF and on a low-end reference VM; else design delta/compression fallback.
SP-4 pip-audit foreign-env mode Correct invocation for auditing an env the tool isn't installed in (freeze-file --no-deps vs newer options). Command line + JSON fixture pinned (P1 feature, but flags decided now).
SP-5 Held-back attribution accuracy On a real numpy/scipy/pandas pin tangle: does graph-based blocker attribution match reality? Preview sentences verified by hand against the tangle; ambiguity fallback path exercised.
SP-6 Windows env discovery sweep PEP 514 registry + py -0p + Microsoft Store Python + uv-managed on a real machine; Store-Python aliasing quirks. Discovery module design note updated; Store Python either supported or explicitly detected-and-explained.

Phase 1 — M1 "Core + CLI" — substantially complete 2026-07-27

Engine trait + both adapters (per SP-1 outcome), probe.py, discovery, index cache + search, plan/resolve/execute two-phase, snapshots, pins, guard graph, error catalog, CLI over all of it. Exit: TESTING L1+L2+L4 green in CI; pipdock update --all survives the SP-5 tangle env with correct summary; dogfood on Kokone's own bot venvs begins.

What works

Every P0 CLI command except health: env list|use, list [--outdated], search, info, install, update, uninstall, pin add|remove|list, snapshot list|create|diff|rollback, doctor, pip-upgrade, engine [pip|uv], index refresh, schema, self report-bug. 173 tests; cargo fmt, clippy -D warnings and npm audit clean.

Verified against real environments, not only unit tests:

  • the held-back conflict from SP-1 is reported and attributed — httpcore 0.15.0 (latest 1.0.9) / httpx 0.23.0 requires httpcore <0.16.0,>=0.15.0 — which no engine produces;
  • update --all resolves the same tangle to latest, applies it, and leaves pip check clean;
  • snapshot rollback restores an environment exactly (PRD §6's "env diff vs snapshot = ∅");
  • the uninstall guard refuses to break requests and exits non-zero, per CLI-SPEC §7;
  • exit codes match CLI-SPEC §5.

Findings that changed the design

plan_requirements Every resolve restates the installed set, or both engines break installed dependents at exit 0 (SP-1).
-U in the dry-run argv Without it pip plans nothing and every package looks held back for no reason. The unit test had encoded the omission instead of the document.
Search ranking Tiered, not score-ordered, and ranked by selection rather than a full sort: 90.5 ms → 16.5 ms against a 50 ms budget.
SnapshotProof --no-snapshot is a named waiver, not an Option, so forgetting a snapshot cannot look like deliberately waiving one.

Not yet done

  • health — belongs to M3 with the tools venv (CODE-HEALTH-SPEC).
  • TESTING L2 in CI — done 2026-07-29. It had run nightly on 07-28 and 07-29 and failed both times; five causes were found and fixed (see L2's first runs below). Run 30460248664 on main is green on all three jobs, including the fixture-drift re-capture that had never passed.
  • TESTING L4 — assert_cmd is a dependency and the clap surface is covered, but the golden-output tests per command are not written.
  • The SP-5 tangle env — done 2026-07-29. A 38-package numpy/scipy/pandas/matplotlib/scikit-learn/statsmodels environment on Python 3.12, with scipy pinned so numpy had to be held back. update --all reported 15 successful, 0 failed, 0 skipped, exited 0, and left pip check clean; numpy stayed at 1.26.4, the scipy pin held (invariant 5), and the snapshot was written before any mutation (invariant 2). See What the dogfood found below — the answer to SP-5's question was "yes, but the explanation was wrong".
  • Settings beyond engine choice — locale, thresholds and the PEP 668 override land with the GUI in M2.

L2's first runs — what actually failed (2026-07-29)

The job ran and failed three times. No cause was a product bug — all five were in the test harness, and all are fixed. Worth reading before touching capture.py or the workflow, because most of them look like the product misbehaving.

1. The rollback assertion targeted the wrong snapshot. The step did snapshot create → update --all → snapshot rollback latest → snapshot diff latest, asserting the prose matches snapshot. But latest moves twice during that flow: update --all writes a Trigger::Plan snapshot before mutating, and snapshot rollback writes a Trigger::Rollback one before restoring (DATA-FLOW §8 — rollback is itself reversible). So the final diff latest compared the restored environment against the pre-rollback state and correctly reported every package as changed. The rollback itself had worked. Fixed by pinning the id from snapshot list --json immediately after create, and asserting structurally on diff --json rather than on prose.

2. fixture-drift could never have gone green. spikes/capture.py built each scenario in a mkdtemp directory whose name carries a random suffix, and redact() replaced only the temp root — so the suffix survived into every committed sidecar, and every re-capture differed. Four further sources of churn were found underneath it: pip's [notice] To update, run: <venv>\Scripts\python.exe (which also leaked the capturing user's home directory into a public repo), CPython object addresses in urllib3 retry warnings, pip's download progress bar with its transfer rate, and uv's per-phase timings (Resolved 3 packages in 775ms). Fixed by redacting all of them in capture.py, splitting the sidecar into meta.json (contract, gated) and capture-provenance.json (versions and argv, excluded from the gate), and capturing cache-free so a warm dev machine and a cold runner agree.

3. The guard step failed on the outcome it was asserting. Fixing the rollback let the job reach the uninstall-guard step, which requires pipdock uninstall urllib3 to exit non-zero (CLI-SPEC §7). The guard did exactly that — but the runner ends a pwsh step with exit $LASTEXITCODE, and nothing had reset it. The other seven end-to-end steps were audited; each ends on a command that legitimately exits 0.

4. The redaction regex ate prose. With the drift diff finally readable, it showed Some things HTTP Core does d<PATH> — pip's --report embeds package descriptions, and does do:\n\n* Sending… reads as drive o: followed by \n path segments. A lookbehind requiring a non-alphanumeric before the drive letter is what distinguishes a path from prose.

5. report-encoding-crash cannot be re-captured on CI. The SP-2 UnicodeEncodeError needs a cp1252 console and a runner's is already UTF-8, so re-capturing there recorded an ordinary successful report and destroyed the evidence. Scenarios now carry a reproducible flag; host-dependent ones are skipped in bulk runs and stay capturable with --only. pip's report also echoes the PEP 508 environment markers, which carry the host OS — normalized, while the interpreter version markers are deliberately left alone, since those can change a resolution.

Verified two ways: locally, both engines' captures run twice back to back differ in 0 of 96 files; on CI, run 30460248664 re-captured against the newest pip and uv and matched the committed fixtures byte for byte.

Two things worth keeping in mind, both recorded in .gitattributes and CLAUDE.md: -text is what preserves the CRLF bytes the fixtures need, and -diff — which was also set — only suppressed the textual diff, so the drift job's own diagnostics could report nothing but "Binary files differ". Faults 3–5 were all found because that was fixed. And the weekly drift job's if: matched the nightly cron, so the PyPI-heavy job the comment says would be "rude" to run daily was running daily; it now has its own '17 4 * * 1' entry.

What the dogfood found (2026-07-29)

SP-5 asked whether graph-based blocker attribution matches reality on a real tangle. It does — and at scale it was also saying things that were not true. The held-back numpy came with eight constraints, four of which do not apply to Python 3.12:

  numpy 1.26.4 (latest 2.5.1)
      pandas 2.1.4 requires numpy <2,>=1.22.4     <- python_version < "3.11"
      pandas 2.1.4 requires numpy <2,>=1.23.2     <- python_version == "3.11"
      pandas 2.1.4 requires numpy <2,>=1.26.0     <- the only pandas branch in force
      statsmodels 0.14.1 requires numpy <2,>=1.22.3   <- python_version == "3.10" and Windows

Requirement has always carried its marker, but only extra == was honoured, so every marker-gated branch of a dependency was reported as though it were in force. That is not just noise — telling someone on 3.12 that "pandas requires numpy >=1.22.4" is false, and ERROR-CATALOG's premise is that what PipDock says about a failure can be trusted. The same bug sat under the uninstall guard, which would refuse a removal on the strength of a dependent that only needs the package on another Python.

Fixed in graph/markers.rs: python_version and python_full_version are evaluated, with and/or/parentheses. Platform markers are left alone — they are constants for a Windows-only v1 and reading them would mean plumbing more out of probe.py for no present gain. An unrecognised marker keeps the requirement, because over-reporting a constraint is noise while dropping one hides the reason a package is stuck. The same environment now reports five constraints, one per package, each genuinely in force.

One thing deliberately not changed: blockers are still derived from the installed set, so a package being upgraded by the very same plan can still be named. Here that meant matplotlib, pandas, scikit-learn and statsmodels appeared even though the plan moves all four to numpy-2-compatible versions, leaving pinned scipy as the only real blocker. Narrowing that needs post-plan metadata the preview does not have, and it makes the sentence over-broad rather than false. Worth revisiting when PdConflictRow is built.

Where to pick up

Pushed to main and green: CI / Rust, CI / Node and CodeQL all pass on the published commit. Three configuration bugs were fixed getting there, each recorded in its own commit message: caller workflows must grant at least what the callee declares (a startup_failure with no logs), cargo audit was pinned to a toolchain too old to build itself, and its informational advisories needed scoping rather than silencing.

Done 2026-07-29, closing out M1's CI debt:

  • Dependabot triage — six PRs resolved. #3 (TypeScript 7) and #4 (rusqlite 0.40) closed per ARCHITECTURE §10; #1, #2, #5, #6 merged. No PRs are open.
  • L2 green — all three jobs, both matrix legs, on main.
  • TESTING L4 — tests/golden.rs, 46 snapshots. See What L4 covers and what it does not below.

Immediate, in rough order:

  1. Repo settings — branch protection, the Discord webhook secrets, and CodeQL/Discussions, per RELEASE-CI §5. None of these are committable and all are owner-only. The updater keypair that used to be on this list is gone: PipDock no longer updates itself (SECURITY §5).

M1 is otherwise closed. Everything else on this list is done, and Stage 1 of M2 — the IPC bridge — is the next piece of work.

What L4 covers and what it does not

TESTING §2 asks for goldens "against a mocked core". There is no seam to mock at — run.rs builds its engine from the global options and drives a real interpreter, and print_preview / print_summary are private functions that println! rather than return. So tests/golden.rs covers what is deterministic without an environment: the clap surface, the exit-code table, the --json error envelope, and schema <T> for all fifteen SCHEMA_TYPES. Preview and summary rendering stays with L2.

Those schema snapshots are also the review surface for M2's wire-format change. They pin two contract bugs as they stand today:

  • ExecutionSummary serializes plan_id / stderr_tail, where DATA-FLOW §6 documents planId / stderrTail
  • Code serializes as its Rust variant name ("EnvInterpreterMissing"), where ERROR-CATALOG documents PD-ENV-001

The error envelope, which main.rs hand-builds rather than deriving, already has the documented shape — so the binary currently disagrees with itself across its two --json paths. Fixing that is Stage 1 work, and the goldens make it a reviewable diff.

When the flow refactor moves rendering out of the CLI ("the flow never prints"), the preview and summary renderers become pure functions of core types and should gain their own snapshots there.

Then M2. The core carries everything the GUI needs; ui/src/ipc fixes the command names, the design tokens and EN/VI catalogs are scaffolded, and the shell renders.

Phase 2 — M2 "GUI shell + core flows" (~7–8 weeks)

Tauri app, design tokens, Environments/Installed/Updates/Search/Pins screens, preview + 3-way conflict UX, console drawer, summary sheet, snapshots UI, legal gate, EN/VI catalogs, settings. Exit: all UI-SPEC click budgets met by manual count; L3 green; VI sweep clean.

The estimate was 3–4 weeks until 2026-07-29, when surveying the code for M2 showed the bridge was scaffolding rather than substance: 26 command names with zero wrappers, one registered Tauri command, 2 of 16 Pd* components, 1 of 5 stores. Scope is unchanged; the estimate moved to match it.

Stage 1 — the IPC bridge — done 2026-07-30

Ten commits, 0ccee04..a553bf7. Everything M2's screens stand on now exists, and there is a running app.

What landed
Wire format #[serde(rename)] on all 30 Code variants so they serialize as PD-*, and rename_all = "camelCase" on every IPC-crossing struct. --json had never matched DATA-FLOW §6 or ERROR-CATALOG §3 — and the two --json paths in the binary disagreed with each other. Held by a Code::ALL round-trip test and a test that walks every SCHEMA_TYPES schema rejecting any property containing _.
Codegen ui/src/ipc/generated.ts from schemars via cargo run -p xtask -- bindings. Deviates from ARCHITECTURE §9's specta; reasoning and the revisit condition are in crates/pipdock-core/src/bindings.rs. Drift is a test failure, so it fires locally as well as in CI.
core::flow UpdateFlow, UninstallFlow, RollbackFlow — resumable state machines, because the GUI's two decision points are IPC round trips with a render in between. plan::execute_rollback is new: core had rollback_plan() and nothing to run it, so the CLI hand-assembled an ExecutionSummary. The flow never prints.
Cancellation CancellationToken through exec.rs, plus kill_on_drop(true) — the 600 s watchdog had never killed anything. A cancelled Phase A does not fall through to Phase B, and a step killed by us is Skipped, not Failed.
Progress ProgressSink carries step/total; scan-progress finally has a producer and a payload (ScanProgress).
Vertical slice Legal gate, Environments, Settings. core::settings, docsHash computed in build.rs, capabilities split and scoped.

Four bugs turned up that were not in the plan, all found by running things rather than reading them: the watchdog leak; --json unparseable because snapshot … written went to stdout; blocker attribution naming constraints from other Pythons (the SP-5 dogfood); and updater:default granted to a plugin whose pubkey was a placeholder.

Stage 2 — Installed + Updates (read-only) — done 2026-08-04

Three PRs, d7da4d7... Both screens render from a real environment, and the slice's exit criteria are met.

What landed
Docs ARCHITECTURE §7's command list rewritten as a table of 32 — it was prose with a|b|c groupings, ambiguous enough that Stage 1 added app_info and plan_decide without amending it. Closes open owner decision #1. DATA-FLOW §7's "List installed" row corrected: it claimed an engine command neither head has ever called.
Core engine::for_id — the CLI held the only copy of engine selection plus a duplicate KEY_ENGINE and a byte-for-byte copy of store::default_app_data. Dist.sizeBytes, summed from RECORD. PkgName validates on deserialize.
Bridge pkg_list, pkg_outdated, pin_list|add|remove, registered and wrapped. pkg_list reads the probe, not the engine; both take a PyEnv so the outdated path never probes.
UI PdBadge, PdEmptyState, PdPackageRow, PdPackageTable (virtualized) — 8 of 16 components now exist. One screen serving both tabs, useEnvStore grown, EN+VI catalogs, Space/Ctrl+A.
L3 The repo's first rendered-component tests, fed from fixtures generated out of the real Rust types by cargo run -p xtask -- ipc-fixtures and held current by a staleness test.

Four bugs turned up that were not in the plan, three of them found by running:

  1. Distribution.files is a 10× trap. The obvious way to sum RECORD took the probe from 551 ms to 5,492 ms on a 352-package environment — on the path the Installed screen runs on every environment open. Parsing RECORD text directly costs 79 ms.
  2. The naive size is wrong for editables, not merely missing. A PEP 660 install has a valid RECORD listing only its import shim: 240 bytes for a package whose sources are 8.1 MiB. Detected via direct_url.json and reported as absent, because a wrong number is worse than none.
  3. A distribution could be listed twice. An editable puts its own .egg-info on sys.path, so importlib.metadata finds it there and at the venv's .dist-info. pip list shows it once. For the table it is an ambiguous join key and two rows a user cannot tell apart.
  4. Every screen fetched twice on mount. React runs effects twice in development and the loadedFor guard is set only after the await, so both runs proceed. env_scan had done this since Stage 1, invisibly, because both scans return the same rows.

Two decisions worth knowing before S3:

  • Row state is three-valued, not two. pkg_list is local, pkg_outdated is networked, so there is a window where outdatedness is unknown. Treating it as "up to date" dims all 200 rows and un-dims a handful a second later — visibly wrong, and untrue while it lasts.
  • "N pinned excluded" is presentation, not enforcement. DATA-FLOW §9.5 is enforced by pins::filter_upgrades at the plan boundary. S3's preview must report the flow's excluded_pins(), not the number the table computed.

Also resolved: UI-SPEC §8's "select-all-visible", which has no meaning under virtualization. It is the current filtered set, and §8 now says so.

Stage 3 — the mutation spine — done 2026-08-04

Two PRs. The app now previews, decides, executes, streams and summarises — PRD G1's promise that every mutating operation is previewed before it runs is real for the first time in the GUI.

What landed
Cancellation A Windows Job Object, so a cancel reaches the build backends python -m pip spawns and not only pip. win32job is a safe wrapper, so unsafe_code = "forbid" stays absolute.
plan-progress A tagged lifecycle (stepStarted / line / stepFinished) instead of a bare line, plus stream. ARCHITECTURE §7 amended.
Wire types FlowStep, Intent, Decision and ProgressEvent cross IPC; roundsRemaining makes MAX_CONFLICT_ROUNDS visible for the first time.
Plan session PlanSlot in AppState holds the resumable flow between the four calls that drive it; PD-RES-003 refuses a second plan.
UI usePlanStore, PdPreviewDiff, PdConflictRow, PdConsoleDrawer, PdSummarySheet — 13 of UI-SPEC §6's 16 components now exist.
L3 The last three of TESTING §2's five obligations, so the list is complete and §3's PR gate means something.

Exit criteria, counted and measured in a running app: "update everything" is exactly 4 clicks — Updates → Select all → Update → Confirm — and it is still 4 with a conflict left at its default, which is §5's second row. Cancelling mid-execution yields a coherent summary with Skipped rows and the cancelled banner. The live region reads "2 of 4 complete" and the drawer groups output by package, both of which were unimplementable before the lifecycle enum existed.

Three specification gaps closed, each recorded in UI-SPEC §4:

  1. Will downgrade has a section. ChangeKind has four variants and §4 named three groups. A compatible resolve routinely moves a package down, and 2.0 → 1.9 under a heading reading "Will upgrade" is misleading about the change most likely to surprise.
  2. Keep compatible is disabled on impossible rows. §4 said every needs-decision row hosts the full 3-way control, but default_decision(is_impossible = true, …) returns Skip — there is no compatible version to keep. The UI mirrors the core rather than offering a choice it would refuse.
  3. The round counter is visible. MAX_CONFLICT_ROUNDS had been in core since M1 with nothing surfacing it, so a user could hit the cap unwarned.

And the third Stage 1 deferral, ExecutionSummary.cancelled copy, which had no specified wording: the summary now says the run stopped part-way, that a package may have been left half-written, and points at the snapshot.

One design change forced by the compiler, worth knowing. UpdateFlow::start took a &Store; Store wraps a rusqlite::Connection and is Send but not Sync, so a future holding one across an await is not Send and a Tauri command cannot return it. It takes the pins instead — one synchronous read at each call site, and the flow no longer touches a database at all.

Stage 4 — search, the dock bay and install — done 2026-08-04

Five commits. Both exit criteria met and measured in the running app. (The "16 of 16 components" claim recorded here was wrong: the count had folded in components that are not on UI-SPEC §6's list. It was 13 of 16 then and after S5 — see §6, which now says so.)

What landed
Index IndexSlot in AppState holds the 864k-name index; warmed on first Search open, never blocking a search. index_search, index_refresh, pkg_metadata.
UI useIndexStore, PdSearch, PdDockBay, PdOfflineBanner, the metadata panel, / to focus.
Install The dock bay resolves through Intent::Install, into the same preview and summary S3 built.

The keystroke budget, measured rather than assumed — and it was being missed.

Median p95 Over 50 ms
First measurement 57.4 ms 79.8 ms 18 of 19
After two fixes 22 ms 24.5 ms 0 of 19

The two fixes were a leading-edge debounce — it was trailing-only, spending 16 ms of a 50 ms budget doing nothing on every keystroke, while the comment above it had claimed leading-edge behaviour since the day it was written — and SEARCH_LIMIT 50 → 20, because React commits every row on every keystroke and the row count is the largest lever on the render half.

A number I reported was wrong, and it mattered. The 613 ms index load was a debug build; in release it is 140 ms, and the worst keystroke is 16 ms rather than 176 ms. The design — hold the index, warm on demand — is unchanged and still right at 140 ms, but it had been justified with a figure four times too large. crates/pipdock-core/tests/search_latency.rs now refuses to run in debug rather than printing a number that means nothing.

Two bugs found by counting clicks, neither reachable by any test written for them:

  1. A plan started from Search resolved into nowhere. PdPlanPanel was owned by PdPackages, so the command ran, the flow parked, and the user saw unchanged search results. A plan belongs to the app — there is one at a time, and which tab is showing is not part of that.
  2. Every result offered [Add], including installed packages, because only PdPackages ever loaded the installed set. That is exactly the mistake DATA-FLOW §4's chips exist to prevent.

Stage 5 — uninstall, the guard dialog and the Pins screen — done 2026-08-09

Twelve commits. Every mutating command in ARCHITECTURE §7 now exists, and removal is the first flow whose whole value is refusing to do what was asked.

What landed
Core GuardReport.breaks carries BrokenDependent { pkg, version, constraint } — the specifier, not just the name. GuardAck makes the guard an enforcement point rather than a dialog. PEP 668 refused on removal. New code PD-RES-004.
Execution execute_uninstall emits the stepStarted/stepFinished lifecycle it never had, checks the cancellation token in its loop, takes a base step index shared with execute_rollback, and honours --no-snapshot.
Bridge One Session enum for all four flows, with typed claims; uninstall_guard, uninstall_execute; plan_cancel async, and now discarding a parked session.
UI PdDialog (the repo's first modal), PdUninstallDialog, PdPinChip, the Pins screen, the row's ✕, and the failed phase.
L3 The first mocked @/ipc, the first store test, the first screen test. 6 files → 9; 59 tests → 79.

Exit criteria, counted and verified in a running app (driven in the Browser pane against a stubbed bridge, per the recipe that found Stage 2's double fetch): uninstall is exactly 3 clicks — Installed → row ✕ → Remove — and 4 when the guard trips, by design. The dialog reads pandas 2.1.4 requires numpy<2,>=1.26.0, naming the dependent and its constraint. Remove dependents too issues a second uninstall_guard over the widened set with no uninstall_execute in between. At the CLI, against a real venv, the guard refuses and exits 2 with every package still installed.

Five live bugs came out of it, none of them S5 features:

  1. A claim that found nothing wedged the slot. plan_decide/plan_execute wrote Busy before discovering there was no plan, and never released — so one out-of-order call from the UI made every later command answer PD-RES-003 for a plan that had never existed.
  2. execute_uninstall emitted no lifecycle markers, and every line carried step: 0. The CLI never noticed because it prints event.line() and nothing else; the console drawer groups on stepStarted and the live region counts stepFinished, so the GUI would have shown an empty drawer against a green suite.
  3. A cancelled removal removed everything, then reported cancelled: true. The loop had no token check at all.
  4. --no-snapshot was parsed and ignored on the uninstall path — the same waiver meant "waive" for an update and nothing for a removal, the one operation with no way back.
  5. The pin button was a tab stop, so Space on it toggled the row's selection instead of pinning. The rule was enforced on the checkbox and forgotten one element later.

And one specification gap closed: DATA-FLOW §9.1 required "a ResolutionReport accepted in this session", which the uninstall path cannot produce — there is nothing to resolve. It now names the proof per plan shape, so the one flow with no preview is no longer the one flow the invariant did not describe.

Stage 6 — snapshots, the env detail and rollback — done 2026-08-09

Eight commits. Every snapshot the app has been writing since S1 is now reachable from it, and P0-8 is complete in both heads.

What landed
Core RollbackPreview and RollbackPlan cross IPC; snapshot::Meta gains a schema rename so embedding it does not mint a second interface. parse_freeze and unrestorable_lines rebuilt on one classifier.
Bridge snapshot_list|create|diff|rollback_preview|rollback. ARCHITECTURE §7's single rollback row becomes two, 32 → 33.
UI PdSnapshotTimeline, PdRollbackPreview, the env detail as a mode of the Environments tab, the diff viewer, and the rollback the summary's copy has promised since S3.
L3 Two more generated fixtures; 9 → 11 test files, 79 → 91 tests. Goldens 52 → 54.

Exit criteria, verified by running. At the CLI against a real venv: create → mutate → snapshot rollback <id> → snapshot diff <id> prints environment matches snapshot — PRD §6's metric, and the timeline afterwards shows the rollback of … entry above the one it restored, which is latest moving twice, live. In the app: rollback = 4 clicks (Open → entry → Rollback… → Roll back), 2 from a run's summary; the preview lists PD-SNP-002 with the verbatim freeze line before the confirm.

Four bugs, three of which no test would have found:

  1. A cancelled rollback killed the package in flight and installed the rest. The restore loop had no token check — S5 gave one to the removal half and not to the install half. Found by writing the cancellation test the plan asked for.
  2. parse_freeze and unrestorable_lines did not partition their input. Two independent filters that disagreed: a line like Foo Bar==1.0 was dropped by both and vanished. In a rollback preview, whose whole job is saying what it cannot put back, that silence is the worst possible failure.
  3. snapshot_list fired twice on open — Stage 2's double fetch, reintroduced exactly, because the guard is written after the await. Invisible in the UI, since both calls return the same rows.
  4. A rollback left the timeline empty. The refetch and the rescan race, and the rescan wiped the result: the snapshot fields were inside NO_PACKAGES, whose reset is keyed to the package slice's environment. The two slices key on different things and now reset separately.

Deliberately not done: DATA-FLOW §8's dry-run resolve of the restore set, which is implemented in neither head. It is a network round trip on an interaction budgeted as instant, and it would resolve against the pre-removal environment — reporting conflicts the two-phase execution never hits. §8 now says so, and the preview's promise is precise instead: it lists what cannot be restored, not everything that might fail. Snapshot retention (snapshot::delete + keep-N) also stays out; a destructive command against the user's only way back needs its own confirm.

Stage 7 — the shell — done 2026-08-09

Eight commits. What the shell was missing was rarely a feature — it was a promise the docs had made and nothing had kept.

What landed
Fonts Inter and JetBrains Mono vendored. --font-sans/--font-mono had named them since the tokens were written and both fell through to a system fallback, because no @font-face was ever registered.
i18n 26 of 32 catalog codes had no copy in either locale. All 32 now do, in locales/*/errors.json, asserted against a generated codes.json — both directions, plus I18N §4's ≤ 90-char cap.
Bug report pipdock_core::report shared by both heads; LogRing in AppState fed by three writers; report_bug_url; the error row's Copy full log and Report bug.
a11y Every --color-* rebound under forced-colors, with borders on the destructive controls whose tint the UA erases. The pseudo-locale, dev-only and verified absent from a production build.
Status line The log toggle does something, and ⚠ n means something.

| Bridge | engine_info, and three tests that make COMMANDS mean something. It listed 32 names while 19 were registered — a wrapper for any of the other 13 typechecked and would have failed at runtime, on a command that looked implemented. | | Keyboard | Focus follows the tab: Ctrl+3 changed the screen and left focus behind, so the next Tab resumed inside the screen the user had just left. |

Deliberately left for later, and named: the Enter primary-action map and folding the six aria-live regions into one — both are cross-screen behaviour that wants the VI sweep and a real screen reader to verify, not a passing test. S7b (Settings' remaining controls) was always a follow-up PR. PdEmptyState now covers every surface that renders a list.

One thing worth keeping: i18next's nsSeparator is :, so I18N §1's seven namespaces would have meant t('errors:PD-NET-001') at every call site — churn across every screen, bought with lazy loading a desktop app that bundles both catalogs cannot use. The app ships one namespace and two files per locale; §1 is amended to say so rather than left describing something that was never built.

M2 is complete. Every P0 feature except Code Health (P0-11) is in both heads, all seven UI-SPEC click budgets are met by hand count, and NOT_YET names the five commands that remain — each against the slice that owes it.

Where to pick up

Phase 3 is complete. P1 (pip upkeep), P2 (the tools venv), P3 (the runners and HealthReport), P4 (the Health screen), P5 (the gated ruff --fix) and P6 (the tail) are all merged, each with a stage table above, and the exit criteria are recorded with the numbers they were measured at. Next is Phase 4 — the release pipeline, the manual charter, and a clean install of the RC on a machine that has never run PipDock.

NOT_YET is down to two, both M3-general: env_add_manual (Browse… has no surface) and logs_tail (needs the logging subsystem). The M3 debt list is otherwise unchanged: dead pins::hold_requirements, unread allow_externally_managed, DATA-FLOW §8's dry-run resolve, snapshot retention, PdEnvSwitcher, CLI-SPEC §6's NDJSON gap, and RunOptions/the confidence floor reaching Settings.

NOT_YET is down to three and only health_fix is Phase 3's. P4 did add two commands after all — health_save_report and, had it been taken, health_cancel — so the earlier claim that it needed none was wrong; both land implemented, so neither is owed.

The four things P4 was warned it would otherwise rediscover all held: @tauri-apps/plugin-dialog really was absent from package.json, the opener really was scoped to GitHub only, UI-SPEC §6 really said 14, and deptry really does name a module rather than a distribution. What the warnings did not cover were the four defects in P4's stage table above, every one of which came from running the thing.

The one open decision is now closed. deptry cannot be told which environment to compare against; the owner chose on 2026-08-13 to keep all four DEP codes and disclose the limitation where findings are shown, rather than --ignore DEP003 (which turns a mislabelled finding into a missing one) or dropping deptry from v1. The screen carries the note under the same condition the CLI does, and the saved Markdown carries it too.

All three Stage 1 deferrals are closed, each in S3 as planned — the plan-progress lifecycle enum, the Windows Job Object, and the ExecutionSummary.cancelled copy. Deferring them to the slice that could verify them worked: each was finished against a running UI or a real process tree rather than against a guess.

Phase 3 — M3 "Health + polish" (~2 weeks)

Tools venv, deptry/vulture/ruff runners + report UI + gated fix, pip upkeep, bug-report deep link, offline states, keyboard map, icon/branding pass. Exit: CODE-HEALTH flows pass on two real projects (one pyproject, one requirements-only).

Decomposed P1 pip upkeep · P2 the tools venv · P3 the runners + HealthReport · P4 the Health screen · P5 the gated ruff --fix. Three of the sentence's items are already spent: the bug-report deep link landed in S7, PdOfflineBanner in S4, and the keyboard map's Ctrl+1..8 in S7 — what remains of "keyboard map" is Enter as primary action and folding the six aria-live regions. The icon/branding pass belongs with Phase 4's release slice; it churns binaries that must not sit inside a feature diff.

Phase 3 · P6 — the tail — done 2026-08-13

Three commits, and Phase 3 is closed. The keyboard item turned out to be hiding a WCAG failure; the rest was owed bookkeeping.

What landed
UI roving tabindex in PdPackageRow (←/→/Esc), Enter as the non-destructive primary, two live regions folded
Core uv::parse_dry_run deleted — the last todo!() in the workspace — and clippy::todo flipped allow → warn
Docs UI-SPEC §8's keyboard map and the two rules behind it; CLI-SPEC §6's stale "health is still the M3 stub"

Pin and Remove were unreachable without a mouse. Every control in a package row is tabIndex={-1}, which is a sound trade for the checkbox — Space does its job from the row — but applied to the two action buttons it removed the actions entirely. The row was the only tab stop and there was nowhere further to go. Found by tabbing to a row in the running app and pressing every key that should have done something; invisible to a suite that clicks.

The audit's other finding is worth keeping: Enter already worked everywhere else, because every other primary action in the app is a real <button> and buttons handle it natively. "Zero Enter handlers" was true and misleading. The gap was one element that only looks like a control.

The live-region pair that mattered was not the one being counted. ROADMAP had said "six regions" for two slices; there were ten. But the duplication that actually reached a user was PdConsoleDrawer and the panel that opens it announcing the same done/total, so a screen reader heard every step twice. Eight now, with both genuine conflicts removed rather than the number merely reduced.

Phase 3 exit criteria — met 2026-08-13

"CODE-HEALTH flows pass on two real projects (one pyproject, one requirements-only)." Run in --release against a warm tools venv:

Project Result Time
requirements-only 1 dependency, 3 dead-code, 3 lint (3 fixable in 1 file) 1.4 s
declares nothing ran: [vulture, ruff], problems: [] — deptry correctly not run 1.3 s
seeded git repo 6 fixable across 2 files; --fix changed exactly those two —

The fix path's refusals were exercised on the same repository: non-TTY without --yes, --yes over a dirty tree, and attrib +R on one target — the last confirming the fix refuses as a whole, leaving the other, writable file untouched.

Still owed, and it needs the real Tauri runtime rather than the stubbed bridge: the OS folder picker, a ruff rule link (to confirm the docs.astral.sh allowlist entry works), and the fix dialog end to end. All three are reachable only from npm run tauri dev.

Phase 3 · P5 — the gated ruff --fix — done 2026-08-13

Five commits. NOT_YET is down to two, both M3-general, and PipDock has its first write outside site-packages and %LOCALAPPDATA%.

What landed
Core health::fix — dirty, FixConsent, consent_ok, ensure_writable, recheck, apply, FixReport; project::validate_project; run::ruff_argv extracted so the re-check cannot drift
Errors PD-PRM-003, internally raised. Code::ALL 33 → 34, and the checklist is nine places, not eight
Bridge health_fix (the last Phase-3 NOT_YET row) and health_dirty
CLI pipdock health --fix, with the --yes rule, the usage error, and a one-document --json
UI the Fix button, one dialog with two states, data-action="fix" in the forced-colors list
Docs DATA-FLOW gains invariant 6 and §9.1's state machine; SECURITY §2 and §8; CODE-HEALTH-SPEC §5; CLI-SPEC §3; UI-SPEC §5 and §7

The invariant question, answered rather than dodged. DATA-FLOW §9.1 and §9.2 are scoped to a mutating engine call, and a ruff fix rewrites a source tree, which no snapshot describes. A snapshot taken here would have no consumer that could use it — invariant 2's own argument for the pip-upkeep exemption, a second time. So: no snapshot, a fifth proof shape (HealthReport + a server-checked FixConsent), and the exemption made visible the way P1's is, in the confirm's own copy.

Three things found by running.

  1. --json --fix printed two documents — the pre-fix report and the FixReport. The first describes a state that no longer exists by the time the command returns, and CLI-SPEC §6 states the contract as one document. Found by reading stdout rather than asserting on it.
  2. A planted git.exe does not win. This is the one place PipDock runs a PATH-resolved program with a user-controlled working directory. Windows' legacy search order would make a binary in a cloned repository run; Rust's Command does its own resolution and skips the current directory. Verified by planting one, not by reasoning about CreateProcess.
  3. The SECURITY paragraph was ahead of the code. Writing "the project folder is validated and a folder inside the tools directory is refused" made it true rather than aspirational: validate_project now refuses it, .. included, and running Health against %LOCALAPPDATA%\PipDock ools answers PD-ENV-003.

Verification, release build, against a seeded repository. 6 fixable across 2 files with an unfixable E722 in a third: after --fix, git diff --stat lists exactly those two files — the third untouched, which is what rules out a stray ruff format — and the run exits 1 with one finding remaining. Non-TTY without --yes refuses (exit 2). A dirty tree with --yes refuses naming the entry count (exit 2) and writes nothing. attrib +R on one target produces PD-PRM-003 naming that file, and the other, perfectly writable, is not written either: the fix refuses as a whole rather than half-rewriting a tree nothing can restore.

Deliberately not done: ruff format. §4's --check toggle stays read-only and default-off; formatting a whole project is a far larger blast radius than fixing seventeen lint findings, and §7's non-goals are contractual.

Phase 3 · P4 — the Health screen — done 2026-08-13

Ten commits, plus a Slice 0 that got the tool pins to their final version first. Code Health is on screen, PdHealthReport is UI-SPEC §6's fifteenth of sixteen, and both halves of Phase 3's exit criterion pass.

What landed
Core health::markdown; reporting(), which keeps deptry out of ran for a project that declares nothing; fixtures::{health_report, health_partial}
Bridge EnvRow.healthProject filled by env_scan and env_probe; health_run persists the folder; health_save_report; env_probe gains a source parameter
UI useHealthStore (a third reset key, and the first that is a pair), PdHealthReport, screens/PdHealth, pickProjectFolder/pickSavePath, apply generalized so two stores share one console reducer
Capabilities docs.astral.sh added to the opener allowlist, dialog:allow-save granted; both recorded in SECURITY §4 and the grant's own description
Copy a health block in both locales, ~30 keys

Four defects found by running, none of them in the plan.

  1. The screen said "no issues found" before anything had run. With a folder remembered and no report yet — the identical lie a tab keyed on ran alone tells about a failed tool, one level up, written by someone who had just spent a commit preventing it. Empty states are the easiest place in this app to claim a result nobody produced.
  2. A project that declares nothing was reported as a deptry failure. deptry raises DependencySpecificationNotFoundError and exits non-zero, which read as PD-HLT-002. PipDock had already detected the state and already told the user results would be limited — then ran a tool that cannot run in it and blamed the tool. Now skipped, and outside ran.
  3. env_probe hardcoded EnvSource::Manual. So Upgrade pip on a registry-discovered interpreter relabelled it Added manually in the chip, and handed that relabelled PyEnv to every later pkg_list. A P1-era defect in the function P4 had to touch anyway.
  4. A test asserted a version Dependabot owns, so merging the ruff bump broke main. Its CI had gone green on a base predating the health module.

Deliberately not done: health-run cancellation. Measured rather than assumed, which is what the plan asked for: a warm three-tool run is 1.3 s and a cold one is that plus the ~15 s tools-venv bootstrap, both far under the 30 s threshold set before the number was known. It is also more than the bridge one-liner it looks like — run_one never passes the sink's token to Command, so wiring set_cancel alone would cancel only the sync, and watchdog() maps a cancel to PD-INT-001, so a cancelled run would render three "PipDock hit an internal error" rows. Worst case remains watchdog-bound at 3 × 120 s; revisit if a real project hits it.

Exit criteria, release build, warm tools venv. A requirements-only project reports 1 dependency, 3 dead-code and 3 lint findings (3 fixable in 1 file) in 1.4 s; a project declaring nothing reports ran: [vulture, ruff] with an empty problems in 1.3 s. In the running app against a stubbed bridge: Ctrl+6 reaches Health, the <h1> takes focus, the remembered folder arrives on the row so the button reads Change folder…, plugin:event|listen crosses before health_run, the three tab counts are 4/3/6 against a fixture carrying 4/3/6, and on the partial report the ruff tab reads did not finish while vulture reads not run and the status line reads ⚠ 1.

Phase 3 · P3 — the runners and HealthReport — done 2026-08-12

Six commits. pipdock health runs deptry, vulture and ruff out of P2's tools venv and reports; health_run streams the same thing over health-progress. NOT_YET is down to three, and only health_fix is still Phase 3's.

What landed
Core health::{project,deptry,vulture,ruff,report,run}; HealthReport and its six sub-types; is_findings_exit; the project folder in health_projects, keyed by env_hash
Bridge health_run, its own Sessions<HealthSession> slot, and forward_progress taking the channel as a &'static str
CLI pipdock health [--path] [--tool …] [--json], exit 1 on findings; --fix refused until P5
Wire format HealthReport in SCHEMA_TYPES — one registration, seven TS declarations, because the generator hoists $defs
Fixtures eight captured documents under tests/fixtures/health/, redacted and -text

Five things the specs had wrong, all found by running the pinned tools.

  1. All three exit non-zero on findings, and disagree about how — deptry 1, ruff 1, vulture 3, with vulture's 1 and 2 being real failures. A plain !out.ok() reports every successful run over a real project as PD-HLT-002, and a suite that only runs them over a clean directory agrees with it.
  2. deptry cannot be told about an environment. §3 said the env is "passed to deptry"; there is no such flag, VIRTUAL_ENV is ignored, and it reads whatever its own interpreter imports. §2's isolation and §3's comparison conflict, isolation wins, and the cost is disclosed rather than papered over.
  3. deptry's --json-output is a file path, not a stream. - would have created a file called - inside the user's project — the exact §1 boundary the flag was chosen to respect.
  4. ruff's docs URL is keyed by rule name, so §6's .../rules/<code> 404s; and it writes findings to stdout with warnings on stderr, which corrupted the first fixture capture.
  5. A tool present but unrunnable reported PD-ENG-001 — "install the engine, or switch engine in Settings", for a corrupted ruff.exe. Now PD-HLT-001, which says re-sync the tools environment, which is the fix.

Exit criteria, run against real projects. pipdock health over a pyproject project reports 1 dependency, 3 dead-code and 3 lint findings and exits 1; over a clean one it exits 0; --tool ruff runs one tool and leaves ran naming only it, so an empty deptry tab is "not run" rather than "clean". Replacing ruff.exe with a text file of the same name — present, so the sync check says Fresh — produced the partial report: deptry and vulture reported normally, ruff landed in problems as PD-HLT-001, and the run still returned a report. Deleting ruff.exe outright instead triggers the implicit sync and repairs it.

Deliberately not done: RunOptions stays off the wire. §4 says the confidence floor comes "from settings", but Settings has three fields and none is this; adding them is a golden, a bindings regeneration and a Settings screen, which is its own slice.

Phase 3 · P1 — pip upkeep — done 2026-08-12

Eight commits. P0-10 is complete in both heads, and three defects older than the slice went with it.

What landed
Core PipEngine::upgrade_pip gains DATA-FLOW §2's PEP 668 guard and an ensurepip fallback; both heads call it directly rather than dispatching on the configured engine. EnvRow.pip_version, read out of the probe's own distribution list.
Bridge pip_upgrade registered — NOT_YET is down to four, two of them M3-general.
UI pip's version on every Environments row; Upgrade pip as a second inline button beside Use, below the planner floor only; a confirm that states the snapshot exemption.
Docs ARCHITECTURE §7's return type corrected; DATA-FLOW §7 (pip unconditionally) and §9.1/§9.2 (a fourth plan shape, and a visible exemption); UI-SPEC §5 gains a row.

Three bugs that were not in the plan, none of them P1's own.

  1. A venv was reported externally managed whenever its base was. A venv has no stdlib, so sysconfig.get_path("stdlib") resolves to the base — and every venv built from a Python shipping the marker (uv-managed, Debian, Homebrew, Fedora) refused every mutation with PD-ENV-002. pip's own check_externally_managed opens with the venv early-return that probe.py was missing. Found by making a venv from a uv-managed 3.11 to test P1 and watching pip-upgrade refuse it.
  2. Every pip below the 22.2 floor is broken on Python 3.12+ — distutils and pkgutil.ImpImporter are both gone — so the button offered exactly when pip is too old ran a pip that could not start, and the traceback's ModuleNotFoundError classified as PD-ENV-003, "could not read this environment". Now falls back to ensurepip, which is stdlib and offline, chosen by asking pip --version rather than matching a traceback that differs by version.
  3. Five locale strings were being discarded by the two-file merge. common.json and errors.json both carry a top-level errors object, and a shallow spread replaced one with the other. errors.unknown was among them — PdErrorRow's fallback for an unrecognized code, the thing S7 built so English developer text could never reach a user, rendering as the literal errors.unknown. Invisible to both existing i18n tests: one checks Code::ALL, the other checks en/vi parity, and both locales lost the same keys.

Exit criteria, counted by hand in the running app. Upgrade pip → Upgrade is 2 clicks from the landing screen. The row's version went 22.1.2 → 26.2.1 with exactly ["pip_upgrade", "env_probe"] crossing the bridge — no env_scan, which is what "without a rescan" means and the only way to tell the two apart. The button then disappeared, the dialog's default focus was Cancel, and a refused upgrade rendered PD-ENV-002 with localized copy and incremented ⚠ n. Against real interpreters: 3.11 with a working pip 22.1.2 upgrades normally; 3.12 with the same pin, which cannot start, is repaired and ends at 26.2.1 running.

Deliberately not done: the Upgrade pip button does not appear merely because a newer pip exists. That needs pkg_outdated — networked, per environment, N calls on the landing screen — to surface something Installed and Updates already offer. What those screens cannot do is upgrade a pip so old the planner behind them refuses to run, which is the only case the button takes.

Phase 3 · P2 — the tools venv — done 2026-08-12

Six commits. Code Health has an environment to run in, and PD-HLT-001's shipped copy — "Re-sync the tools environment" — now refers to something the user can actually do.

What landed
Core health::pins/pins_hash/requirements_body over tools/tools-requirements.txt, which had existed since the first scaffold commit and which nothing read. sync_tools_venv(tools_dir, base_python, sink) replaces the module's todo!(): create, install, verify. ToolsManifest + SyncNeed + needs_sync, choose_tools_python.
Errors PD-HLT-004 (HltVenvCreateFailed), the eight-place checklist in one commit. Code::ALL 32 → 33.
CLI pipdock tools sync [--force] [--python] and tools status; doctor reports whether Health can run at all, without folding it into its exit rule.
Wire format None. No Tauri command, no NOT_YET edit, no COMMANDS entry, no new SCHEMA_TYPES. P2 ships the health-progress producer; P3 ships the channel, when health_run exists to emit on it.
Docs CODE-HEALTH-SPEC §2 amended on four points; CLI-SPEC §3 gains the commands and §6 gains a recorded, unclosed NDJSON gap.

Five findings that were not in the plan.

  1. TOOLS_PYTHON_MAX should not exist, and the premise for it was wrong. ROADMAP called it "the top unknown in all of M3" on the grounds that deptry ships compiled wheels with no cp314 build. deptry ships a Rust extension built against CPython's stable ABI — one cp310-abi3 wheel, installed here on both 3.12.10 and 3.14.6. No ceiling was added; --only-binary=:all: makes a future gap a clean PD-NET-011 instead of an sdist build.
  2. tools/tools-requirements.txt already existed and was already Dependabot-wired. P2 consumes it rather than creating it, which deleted about half a day of the estimate.
  3. A re-sync could not repair the one state it exists to repair. pip install -r over satisfied pins is a no-op, so replacing a deleted ruff.exe installed nothing and then failed verification with PD-HLT-001 — and wedged, because the manifest had already been deleted. The venv is rebuilt now, not repaired. Found by deleting ruff.exe; invisible to a suite that only ever synced into an empty directory.
  4. SyncNeed::ToolMissing(String) could not be serialized. #[serde(tag = "state")] is internal tagging and serde cannot represent a tagged newtype. It passed every unit test that constructed it and panicked the first time tools status --json met a missing tool. Now a struct variant, with a test that serializes all five states.
  5. The pin closure is not ABI-free, and the reason to exclude pip-audit is narrower than assumed. deptry pulls tomli, which ships per-CPython wheels — but also a py3-none-any fallback, so a missing cpXXX build degrades to pure Python. pip-audit's closure contains msgpack, which publishes no universal fallback and would hard-fail under --only-binary=:all:. That, not "exposure is zero", is why the ledger's fourth pin stays out of the venv.

Exit criteria, measured by running it (debug build, i7-14700KF, Python 3.14.6): cold bootstrap 15.2 s, warm re-run 31 ms, repair-after-quarantine 13.3 s. PIP_NO_INDEX=1 gives error[PD-NET-011], exit 6, an empty venv and — the thing that matters — no manifest.json left behind, so the next online run does not believe it is fresh. A forced --python onto 3.12.10 installs the same abi3 wheel. A sync killed mid-download leaves zero orphan pip.exe and no manifest.

Deliberately not done: deleting the dead engine::uv::parse_dry_run and flipping clippy::todo from allow to warn. After this slice that function holds the last todo!() in the workspace, so the flip is a clean two-line follow-up — but not inside a feature diff. sync_tools_venv also takes no --break-system-packages path and needs none: the tools venv is PipDock's own and is never PEP 668 managed.

Phase 4 — RC → v1.0 (~1–2 weeks)

Release pipeline live (bundling, checksums), manual charter executed, docs/README screenshots, legal files public, a clean install of the RC verified on a machine that has never run PipDock. Exit: RELEASE-CI §5 checklist fully checked; v1.0.0 published + notify fan-out.

Post-1.0 (P1 wave)

Security tab (pip-audit), pin auto-suggest, requirements/constraints export-import, cache manager, command palette, dependency graph view, scheduled check. Then P2 candidates by demand: macOS/Linux, per-env engine, elevation broker, winget, JP locale.

Standing risks

Risk Mitigation
uv output churn between releases weekly latest-engine parser CI job (TESTING L1) + PD-ENG-003 graceful degrade + "switch to pip" escape hatch
pip UX flags deprecations (e.g. report format changes) same job covers pip; adapter version-gates per engine version
Index size growth SP-3 fallback design kept in the drawer
Scope creep from Health into a linter IDE CODE-HEALTH §7 non-goals are contractual