bullswarm 0.28.0 → 0.28.2

package/CHANGELOG.md

@@ -1,5 +1,149 @@
 # bullswarm changelog
 
+## 0.28.2 — terminal glyph fallback, issue watcher, health on a fresh home
+
+- `bullswarm workflow` repainted flashing `?` characters on macOS
+  Terminal.app. Cause is font coverage, not encoding: parsing the `cmap` of
+  every monospace face Terminal.app offers shows **zero** glyphs in the
+  Braille block `U+2800-U+28FF` — Andale Mono, Menlo, SF Mono, Monaco and
+  Courier New are all 0/256 — and none of the five has `⧖` (`U+29D6`) or `⟡`
+  (`U+27E1`). The dashboard's 10-frame Braille spinner advances every 400ms,
+  so every repaint drew a question mark that flashed; the static `✓ ✗ ⊘ ◇ ↳
+  ⚠ ● ○ ■ ♡ ◆ ↻ ↺ ▶ ⚙` icons beside it were the steady ones. Terminal.app
+  does not substitute another face for these ranges.
+- New `src/lib/glyphs.js` holds one table per mode and the detection order:
+  `BULLSWARM_ASCII` forces ASCII, `BULLSWARM_UNICODE` forces Unicode and
+  overrides detection, then a non-UTF-8 locale, `TERM=dumb`/`TERM=linux`, and
+  `TERM_PROGRAM=Apple_Terminal` select ASCII. Every replacement is one column
+  of printable ASCII, because panel width maths uses `String#length` and a
+  wide substitute would shear the borders. Box drawing, `· › × … ≈ ≥ █` are in
+  all five fonts and are left alone, so panels keep their borders.
+- Wired through the three live-refreshing renderers only:
+  `src/workflow/dashboard.js`, `src/workflow/watch-cli.js` and
+  `src/strategy-dashboard.js`. One-shot `✓`/`✗` message prefixes in
+  `src/workflow/cli.js`, `src/help.js`, `src/integrate.js`,
+  `src/workflow/runs-cli.js`, `src/strategy-cli.js` and
+  `src/lib/cli-flags.js` are untouched — they do not flash and are a separate
+  change.
+- `spinnerGlyph` now parks a non-finite frame counter on frame 0. The previous
+  expression `SPINNER_FRAMES[Math.abs(Number(f) || 0) % SPINNER_FRAMES.length]`
+  returned `undefined` for `Infinity`, so a repaint could have painted the
+  string "undefined" into the frame.
+- `tests/glyphs.test.js` covers the detection matrix (including that `''`,
+  `'0'` and `'false'` do not switch tables, and that `LC_ALL` outranks
+  `LANG`), asserts every Unicode entry has a one-column pure-ASCII twin that
+  is never `?`, and renders both dashboards, `renderWatchSnapshot`,
+  `renderWatchEvent` over 15 event shapes and `renderAnalysisProgress` at five
+  spinner frames, asserting no substituted glyph survives in ASCII mode while
+  the Braille spinner still appears in Unicode mode.
+- `tests/strategy-cli.test.js`, `tests/workflow-dashboard.test.js` and
+  `tests/workflow-watch.test.js` assert the Unicode presentation, so they now
+  pin `BULLSWARM_UNICODE=1` and `delete process.env.BULLSWARM_ASCII`; without
+  the pin the suite's own output depended on which terminal the developer ran
+  it in, and without the delete an ambient `BULLSWARM_ASCII=1` — the very
+  workaround the README hands an affected user — outranked the pin and broke
+  22 tests.
+- The four arrows `↑ ↓ ← →` are deliberately not substituted. They are only
+  missing from Monaco, and they live in the frozen `DASHBOARD_KEYS` constant,
+  which binds at import time — before any env pin could apply.
+- `scripts/issue-watcher/` — a durable GitHub issue watcher for this
+  repository, installed as a launchd agent
+  (`node scripts/issue-watcher/install.mjs`). Idle costs ZERO model tokens:
+  a pass is one `gh issue list` and nothing else. Only when an issue appears
+  that is new since install does it delegate — one
+  `bullswarm run --lane analyze` to triage (label + plain-words comment),
+  and for a bug it judged fixable at confidence >= 0.7 one
+  `bullswarm run --lane build` to fix, which the watcher then verifies
+  itself (test suite green in its own clone, diff non-empty) before pushing
+  `fix/issue-<n>` and opening a pull request. Both delegations run on the
+  owner's subscriptions through normal bullswarm routing, never API tokens.
+  Capped at 6 triages and 2 fix attempts per UTC day, one fix attempt per
+  issue; guarded by a pass lock, a `paused` file, and a 3-attempt triage
+  retry cap. Delegates get no GitHub credentials and issue text is quoted as
+  untrusted data. It never merges, closes, or releases — a maintainer does.
+
+## 0.28.1 — summary bytes on the wire, monthly pacing
+
+- `workflow runs result <id> --summary` was budgeted by `fitResultSummary`
+  against compact `JSON.stringify(summary)` (`RESULT_SUMMARY_BYTE_BUDGET =
+  4096` in `src/workflow/v2-outcome.js`) but `jsonOut` printed
+  `JSON.stringify(obj, null, 2)`, so the bytes on the wire exceeded the
+  budget. `--summary` now prints compact single-line JSON
+  (`JSON.stringify(obj)` in `src/workflow/runs-cli.js`); `--json` without
+  `--summary` still pretty-prints the full envelope, and `--summary --json`
+  stays identical to `--summary`. Measured on
+  `tests/fixtures/real-result-ze5xz2.json` through the summariser / CLI:
+  compact `--summary` is 3,786 bytes
+  (`tests/workflow-result-summary.test.js` prints `result-summary size:
+  full=57141 summary=3786`; same figure as
+  `Buffer.byteLength(JSON.stringify(summarizeV2Result(fixture)))`); the
+  full envelope as `--json` prints it is 60,709 bytes
+  (`JSON.stringify(envelope, null, 2)` plus the trailing newline
+  `console.log` adds — `tests/workflow-result-summary.test.js` prints
+  `result-summary cli: prettyFull=60709`).
+
+- Routing paced every pool by `windows.seven_day ?? windows.monthly`
+  (`paceSnapshot` in `src/meters/framework.js`), so command-code — whose real
+  budget is a monthly credit allocation — was paced by its weekly rate-limit
+  window. Live meter, captured 2026-09-09T10:45:28Z and evaluated at
+  2026-09-09T11:09:44.982Z: weekly `used 73.1% elapsed 91.8% surplus +18.7`
+  against monthly `used 79.4% elapsed 75.3% surplus -4.1`, with 14.43 of 70
+  credits left for the 7.66 days to the 2026-09-17T03:06:55Z reset. Routing
+  therefore called it "the most-behind capable pool" and kept sending it work
+  while its monthly budget was already 4.1 points overspent. Pacing is now per
+  pool: `pacingWindowFor({connector, subscription})` resolves
+  `state.strategy.subscriptions[pool].quotaWindow`, then
+  `connector.subscription.quotaWindow`, normalised to `weekly` | `monthly` |
+  `null` (any other label — including a pre-0.28.1 free-text one — is ignored
+  for pacing and keeps the old weekly-first order). `paceSnapshot(snapshot,
+  nowMs, {pacingWindow})` takes `monthly` → `windows.monthly ??
+  windows.seven_day`, `weekly`/`null` → `windows.seven_day ??
+  windows.monthly`, and returns `pacingWindow` naming the window actually
+  used. `src/lib/config.js` `buildPools` resolves the choice once per pool
+  (where the connector and the state both are) and re-paces `usedPct`,
+  `elapsedPct`, `pace` and `paceResetsAt` off `reading.windows`, so cache,
+  stale and live readings are paced identically; the pool view carries
+  `pacingWindow`. The 5h gate is untouched: `command-code` still gates on 5h
+  `25.2%`. Connectors already declared this — `command-code` and the kaihk
+  pools `monthly`, `claude-code`/`codex`/`grok` `weekly`; nothing read it for
+  pacing before.
+- The spend model follows the pacing window. `WINDOW_KEYS` (framework.js)
+  gains `monthly: {snapshot: 'monthly', history: 'monthly', windowMs: null}`,
+  and `rateForWindow` (`src/lib/spend.js`) derives the bootstrap window start
+  with `meta.windowMs ?? monthlyWindowMs(resetsAtMs)` — the calendar month
+  ending at the provider's `resets_at` (M2), never an assumed 30 days.
+  `attachSpend` now writes `pool.spend.monthly` and `pool.projectedMonthlyPct`
+  next to the fiveHour/weekly fields, plus `pool.spend.pacing = {window,
+  ratePerMinute, source, samples}` and `pool.projectedPacingPct` for the
+  window that paces the pool (default `weekly`, so a pool that declares
+  nothing keeps its old numbers). `inflightLoad` (`src/lib/route.js`) charges
+  the in-flight penalty from `spend.pacing?.ratePerMinute ??
+  spend.weekly?.ratePerMinute` with that rate's own source label, so the
+  surplus and the penalty are measured in the same window; candidate rows gain
+  `pacingWindow` and `projectedPacingPct` beside the unchanged
+  `projectedWeeklyPct`.
+- Operator control and display. `bullswarm strategy set-subscription <pool>
+  --quota-window <weekly|monthly>` now selects the window that paces routing
+  (help text in `src/help.js`) and validates it: anything else exits 2 with
+  `--quota-window must be weekly or monthly (or unknown to clear)`, and
+  `unknown` clears the override back to the connector's declaration. Labels
+  already stored are ignored for pacing, never rejected on read. `bullswarm
+  pools` names the window in the meter column — `cmd-fixture    cost=5
+  lanes=chore monthly used 79.4% elapsed 75.3% [cache] surplus=-4.1
+  inflight=0 5h=25.2% ready` — and `pools --json` entries carry
+  `pacingWindow`. `strategy refresh`/`show` print the window on each
+  subscription line (`command-code: GOAT · ... · monthly 79.4% used · surplus
+  -4.1`) and carry `pacingWindow` next to the free-text `quotaWindow` label in
+  `--json`; `strategy inventory --json` carries it per provider.
+- Tests: 738 -> 750, 0 failures. The new behaviour is covered in
+  `tests/meters.test.js` (the live command-code snapshot as a fixture, the
+  helper's precedence, `buildPools` pacing), `tests/spend.test.js` (monthly
+  bootstrap window start, `spend.pacing`/`projectedPacingPct`),
+  `tests/route.test.js` (the penalty on the pacing window; weekly-only pools
+  unchanged), `tests/strategy-cli.test.js` (`--quota-window` validation) and
+  `tests/assignments.test.js` (the `pools` meter column and `--json`
+  `pacingWindow`).
+
 ## 0.28.0 — context diet
 
 - `workflow runs result <id> --summary` prints a compact status-loop
@@ -77,9 +221,10 @@
   39,288; compact `JSON.stringify` of the parsed envelope is 57,141.
   `summarizeV2Result` of that fixture is 3,786 bytes —
   `tests/workflow-result-summary.test.js` prints `result-summary size:
-  full=57141 summary=3786`. The 0.28.0 goal recorded the integrator's
-  inputs as 60,790 bytes; `wc -c` of the seven dependency out-files under
-  `.diet-inputs/` sums to 46,022 (out-surface 18,659, out-routing-cleanup
+  full=57141 summary=3786` (numbers re-measured in 0.28.1). The 0.28.0
+  goal recorded the integrator's inputs as 60,790 bytes; `wc -c` of the
+  seven dependency out-files under `.diet-inputs/` sums to 46,022
+  (out-surface 18,659, out-routing-cleanup
   10,202, out-state-bugs 7,052, out-docs 6,831, out-dead-kernel 1,377,
   out-verify-gate 974, out-dead-code 927) and the integrator task file is
   14,768 (`wc -c .diet-inputs/task-integrate-attempt-1.md`), which

package/README.md

@@ -51,7 +51,12 @@ detaches safely.
    passing verification.
 2. **Pace by meter.** The scheduling resource is the subscription window:
    elapsed% minus used%, most-behind pool wins. Pace may only promote a
-   *cheaper* pool. Lanes are work-nature, never hard-coded to pools. The
+   *cheaper* pool. Lanes are work-nature, never hard-coded to pools. Which
+   window paces one pool is the subscription window that pool's connector
+   declares (`quotaWindow`: weekly for claude-code, codex and grok; monthly
+   for command-code and the kaihk pools), overridable per pool with
+   `bullswarm strategy set-subscription <pool> --quota-window <weekly|monthly>`
+   — `bullswarm pools` names it in the meter column. The
    5-hour window never paces — it gates: a pool at or above 75% of it is
    chosen only when no eligible pool below that line exists, and one at or
    above 90% is not dispatched at all.
@@ -656,6 +661,10 @@ These are UTF-8 byte counts, never tokens.
 requirement as `{ id, status, mandatory, evidenceCount, why }`, each action as
 `{ id, kind, lane, effort, status, pool, model, reasoning, wallSec, outFile,
 bytes }`, `concerns: { count, first }`, `usage`, and `next: { full, runDir, outputs }` — every output name is a basename inside `next.runDir`.
+`--summary` is single-line JSON (`JSON.stringify`), so the bytes on the wire
+match the 4,096-byte fitter budget. As printed by the CLI on
+`tests/fixtures/real-result-ze5xz2.json`, the compact summary is 3,786 bytes
+and the pretty full envelope (`--json` alone) is 60,709 bytes.
 The full `bullswarm.workflow.result.v2` envelope is unchanged and remains the
 default. Read it (`--json` alone) on a failed or partial run, or before judging
 evidence. A terminal `workflow watch` prints the same compact command as
@@ -791,6 +800,25 @@ to toggle Timeline and Phases, then use Enter/Esc for agents and activity.
 bullswarm workflow tui
 ```
 
+#### Terminal glyphs
+
+The live views draw a Braille spinner and symbol status icons. macOS
+Terminal.app cannot render them: Andale Mono, Menlo, SF Mono, Monaco and
+Courier New all have zero glyphs in `U+2800-U+28FF`, and none has `⧖`, so the
+dashboard repaints a flashing `?` where each one should be. Apple Terminal is
+detected and given a one-column ASCII table instead (`|/-\` spinner, `+`
+succeeded, `x` failed, `:` waiting, `#` blocked). Panel borders are unchanged —
+box drawing is present in every one of those fonts.
+
+Override the detection either way:
+
+```bash
+BULLSWARM_ASCII=1 bullswarm workflow      # force ascii (any terminal showing ?)
+BULLSWARM_UNICODE=1 bullswarm workflow    # force unicode (font does have them)
+```
+
+A non-UTF-8 locale, `TERM=dumb` and `TERM=linux` also select ASCII.
+
 The same dashboard is agent-friendly and never requires a TTY when used with
 JSON/control flags:
 
@@ -879,6 +907,19 @@ runtime value and uses its matching connector rate metadata for the attempt's
 cost estimate. Unknown or provider-hidden model identity remains explicitly
 unknown.
 
+## Repository operations
+
+`scripts/issue-watcher/` is a launchd agent that watches this repository's
+GitHub issues. Idle it costs zero model tokens — a pass is one `gh issue
+list` and nothing else. When an issue arrives that is new since install, it
+delegates one analyze run to triage it (label plus a plain-words comment),
+and for a bug it judged fixable at confidence >= 0.7 one build run to fix
+it, verifying the result itself (suite green in its own clone, diff
+non-empty) before pushing `fix/issue-<n>` and opening a pull request. Both
+delegations route through `bullswarm run` on the owner's subscriptions. It
+never merges, closes, or releases. Install, guards, cost model and paths:
+[`scripts/issue-watcher/README.md`](scripts/issue-watcher/README.md).
+
 ## The verdict
 
 ```json