- Added `recoverOrphanedBackups` to promote `.jsonl..bak` files back to their primary path when the primary is missing, preventing data loss after a mid-rename crash.
- Changed backup filename from dot-prefixed to plain `..bak` so the shared `*.bak` glob can find it on both real and in-memory storage backends.
- Surfaced the original EPERM as the error `cause` and included both original and retry messages when rollback also fails.
- Replaced baked module-load value with per-call `isWebPExcluded()` so runtime env changes take effect.
- Only `"1"` and `"true"` (case-insensitive) enable exclusion; empty string and `"0"` are treated as disabled.
- Fast path now bypassed for WebP sources when exclusion is active.
- Explicit error surfaced when decode fails and WebP exclusion cannot be honored.
Anchors are formatted by read/search as LINE+HASH|TEXT, and lines may be
prefixed with marker decoration (*, >, +, -). The parser previously required
a bare LINE+HASH and rejected verbatim copy-pasted anchors with:
line N: expected a full anchor such as "119sr", ...; got "364sp|".
Loosen LID_CAPTURE_RE to allow optional leading decoration and an optional
trailing |... body on each anchor (including each side of a range).
- Added `retry.maxDelayMs` to the settings schema and interfaces, with a default cap for provider backoff delays.
- Updated session auto-retry logic to fail fast when a requested wait exceeds the cap without fallback, emitting terminal auto-retry failure state.
- Propagated retry state and failure data into task progress and rendering so children show retry/wait details and reminder prompts stop after terminal errors.
Separated model selector provider tab labels from provider ids so human-readable labels like Ollama Cloud refresh and filter the underlying ollama-cloud models.
Fixes#1153
In a compiled binary, Bun.resolveSync(spec, import.meta.dir) throws
'Cannot find module' because import.meta.dir is inside /$bunfs/root
and the virtual FS exposes no node_modules tree at runtime.
Previously this throw propagated through rewriteLegacyPiImports ->
rewriteLegacyPiImportsForRuntime -> mirrorLegacyPiFile ->
loadLegacyPiModule -> loadExtension, which swallowed it as 'Failed to
load extension' and silently dropped any plugin whose files imported
@mariozechner/pi-ai (or any @mariozechner/pi-* whose bundled
counterpart isn't reachable via resolveSync in the binary).
Fix: wrap the resolution call in rewriteLegacyPiImports in a try/catch
and return the original match on failure. rewriteBareImportsForLegacyExtension
runs immediately afterwards in every call path and already resolves bare
specifiers against the importer's real filesystem directory, so it picks
up @mariozechner/pi-ai from the plugin's installed peer deps instead.
Apply the same fallback to resolveLegacyPiSpecifier (the Bun plugin
shim's onResolve handler) for tool/hook files loaded directly via Bun's
import system rather than through loadLegacyPiModule.
Fixes#1215
- Removed the exported formatBashFixupNotice helper from bash command fixup utilities.
- Removed BashTool's one-time bash-fixup notice tracking and stopped emitting those notices when fixups were applied.
Root cause (verified on user's environment):
- User commit `296641213` swapped status-line's context% computation from cheap `calculatePromptTokens(lastAssistantMessage.usage)` to `computeContextBreakdown(session)`, which walks EVERY message and runs native `countTokens` (~0.5 ms per message).
- The 2-second TTL cache helps for steady-state idle but every cache MISS is a full sweep.
- `updateEditorTopBorder()` is invoked on EVERY agent event (event-controller.ts:163 — `agent_start`, `delta`, `agent_end`, `tool_*`). Each delta during streaming can trigger a cache miss.
- User session has 2,312 messages → each full sweep is ~1,120 ms blocking.
- During streaming the UI freezes for ~1.1 s every ~2 s, producing the user-visible 'jittery rendering' ("버벅거림") and 'status bar disappearing' symptoms.
Fix:
`StatusLineComponent.getCachedContextBreakdown()` (renamed from `#getCachedContextBreakdown` so unit tests can exercise it directly) now uses an incremental per-message token cache that exploits the append-only nature of `session.messages`:
1. Message tokens (the dominant cost): cached per-index. New messages are tokenized as they arrive; previously-cached messages are reused. The LAST message is always recomputed because its content may still be growing during streaming. Compaction (messages.length shrinks) resets the cache.
2. Non-message tokens (system prompt + tools + skills): cached separately, invalidated only when a cheap inputs-identity fingerprint changes (model swap, skill toggle, tool registration). These rarely change during a session.
Required exposing three helpers from `modes/utils/context-usage.ts` (`estimateSkillsTokens`, `estimateToolSchemaTokens`, `computeNonMessageTokens`) so the status-line cache can call them directly.
Performance (2,300-message synthetic session, measured on user's M-series Mac):
- COLD warm-up call: ~75 ms (one-time, runs at OMP startup before any streaming)
- WARM refresh, no new message: ~0.04 ms (20 calls = 0.7 ms total)
- WARM refresh, 1 new message: ~0.02 ms
vs. prior implementation:
- Per cache-miss call: ~1,120 ms blocking
- 28,000× speedup on warm-state refresh
`computeContextBreakdown` itself is untouched — `/context` slash command continues to use it, and its output matches the status-line context% for the same session state (parity preserved).
Tests: 6 new cases in `packages/coding-agent/test/status-line-context-cache.test.ts` covering cold/warm/append/compaction/non-message-invalidation/zero-messages and a perf smoke test asserting 20 warm refreshes on a 200-message session complete in <100 ms.
Full suite: 3,199 tests, 26 pre-existing failures (status-line accent / log_experiment timing-flaky / skills / github tool / workspace-tree / tool path — all unrelated and baseline-confirmed). Lint: 1 pre-existing import-order issue in `event-controller-plan-ready.test.ts` unchanged.
Status-line's context_pct segment was computing tokens via
calculatePromptTokens(lastAssistantMessage.usage), which sums input +
cacheRead + cacheWrite from the Anthropic API usage object. The /context
slash command is computed by computeContextBreakdown, an offline estimate
over the live session state (systemPrompt + tools + skills + messages).
Both numbers are correct under their own definition, but they can
diverge by 2x+ on the same session when a turn rotates cache tiers
(e.g. 5m → 1h ephemeral re-cache) and cache_creation_input_tokens spikes.
Users read the two surfaces as one consistent dashboard and treat the
mismatch as a bug.
Repro: same session at the same moment reports 212K (21.2%) in /context
and 44.2%/1M in the status line — ~230K gap driven by per-turn
cache_creation on a system-prompt boundary.
This change makes status-line use the same computeContextBreakdown
source as /context so both surfaces stay consistent. The breakdown
result is cached with a 2s TTL inside the component so the per-frame
status-line render does not re-walk every message via
estimateMessagesTokens on long sessions. The Anthropic API per-turn
prompt size remains observable via existing token_in / cache_read /
cache_write / token_total segments.
- New 'usage' status-line segment showing Anthropic 5h/7d quota
- Background refresh (5min TTL) via fetchUsageReports
- thinking.max symbol added to UNICODE_SYMBOLS
Personal patch consolidated into branch.
Bug: Ctrl+C on the ask tool selector threw ToolAbortError, the turn
ended with stopReason === "aborted", and handleBackgroundEvent fired
sendCompletionNotification() unconditionally — producing a misleading
"Task complete" desktop toast for a turn that never actually completed.
Fix mirrors the stopReason filter already used by
#currentContextTokens, #handleMessageEnd, and the retry / TTSR /
compaction skip paths across agent-session.ts: check the most recent
assistant message via session.getLastAssistantMessage() and return
early when stopReason is "aborted" or "error".
Test coverage (event-controller-abort-guard.test.ts, 6 cases):
- aborted -> 0 sendNotification calls
- error -> 0 calls
- stop -> 1 call (normal completion)
- no last assistant message -> proceeds (defensive)
- isBackgrounded=false (foreground) -> still 0
- completion.notify=off -> still 0
Matching guard applied to the standalone desktop-notify extension
(~/.omp/agent/extensions/desktop-notify/index.ts) which is currently
the live producer of completion toasts after Phase 1 of
seed_0ca7e1143ac1.
ImmutablePrefix caches system prompt + tool specs after first build()
so subsequent turns reuse identical byte sequences. AppendOnlyLog
converts messages once via syncMessages() and only appends deltas
on further turns — prior-turn bytes stay stable.
- New module: packages/agent/src/append-only-context.ts
StablePrefix, AppendOnlyLog, AppendOnlyContextManager
- AppendOnlyContextManager added to AgentLoopConfig
- Wired into streamAssistantResponse in agent-loop.ts
- Toggleable via provider.appendOnlyContext setting (auto/on/off)
- Default auto enables for deepseek provider
- 38 tests covering prefix, log, sync, compaction handling
- /session info surfaces current active state
Replaced overwrite-style session rewrites with an EPERM fallback that moves the old session file aside before retrying and restores it if the retry fails.
Added regression coverage for active-session rewrite recovery so the session remains writable after the fallback.
Fixes#1337
Adds optional autoloadSkills field to agent frontmatter that automatically loads listed skills when a sub-agent is spawned. Uses the same buildSkillPromptMessage + sendCustomMessage mechanism as interactive skill loading, queued via sendCustomMessage({ triggerTurn: false }) before the first session.prompt(task). No extra agent turns, no new injection path. Skills stay in listing for sub-resource access. Compaction behavior matches manual loading. Unknown skill names silently skipped.
Lore-id: f85fdbdc
Constraint: autoload must use buildSkillPromptMessage + sendCustomMessage, never modify systemPrompt or use contextFiles
Constraint: triggerTurn must be false to avoid extra agent turns
Rejected: append to systemPrompt | agent cannot distinguish skill content from own instructions
Rejected: contextFiles injection | agent sees opaque file blob, cannot discover sub-resources
Rejected: promptCustomMessage per skill | N extra agent turns with model inference
Directive: autoload skill names are resolved against parent session skill list at spawn time in task/index.ts
Tested: TypeScript compiles clean with tsc --noEmit
Tested: parseAgentFields parses array and CSV string frontmatter
Tested: parseAgentFields returns undefined for absent and empty fields
Not-tested: bun test cannot run locally due to missing pi_natives native addon (requires Rust toolchain)
Confidence: high
Scope-risk: moderate
Reversibility: clean
Keep chat notifications emitted during session_start visible after the initial transcript render rebuilds the chat container. Add regression coverage for preserving startup notifications during initial render.
Fixes#1316
Gated WebP encoding behind OMP_NO_WEBP environment variable so that local
llama.cpp vision models (which use the STB library that lacks WebP
decoding support) can accept browser snapshots without returning HTTP 400.
Upstream reference: cline/cline PR #9837
(https://github.com/cline/cline/pull/9837) implements a similar workaround
for the same llama.cpp STB WebP incompatibility.
When /plan <text> or /goal set <text> is typed while plan/goal mode is
already active, the handler shows a warning and clears the editor, but
the typed text was silently discarded — the user could not recover it
with Up Arrow.
Fix: save the text to editor history before clearing, so it remains
recoverable via input history.
Fixes#1287
When session.prompt() returns, idle-flush tasks for async-job result
deliveries are scheduled via #schedulePostPromptTask (1ms delay) and
added to #postPromptTasks immediately. The 800ms loop timer could fire
in that window before isStreaming became true, causing the loop to
submit the next prompt while the delivery turn was still pending. The
delivery then hit AgentBusyError and the job result was silently dropped.
Add AgentSession.hasPostPromptWork (= #postPromptTasks.size > 0) and
include it in #isLoopAutoSubmitBlocked() alongside isStreaming and
isCompacting. Add a regression test that verifies the loop defers when
hasPostPromptWork is true and fires once it becomes false.
Fixes#1294
Allowed /goal set to replace the current active goal instead of rejecting and discarding the command input.
Added goal runtime and interactive-mode regression coverage for active replacements.
Fixes#1293
WSLg exposes WAYLAND_DISPLAY, so readImageFromClipboard() took the
native arboard path on WSL2. arboard::Clipboard::get_image() returns
ContentNotAvailable on WSLg because the Wayland clipboard does not
carry image payloads from the Windows clipboard, and that surfaced as
silent 'No image in clipboard' on Ctrl+V.
Detect WSL via WSL_DISTRO_NAME / WSL_INTEROP and read the image with a
PowerShell one-liner that emits base64-encoded PNG bytes from
[System.Windows.Forms.Clipboard]::GetImage(). Fall back to the native
bridge when PowerShell returns nothing, exits non-zero, or is missing,
so non-WSLg Wayland setups continue working unchanged.
Fixes#1280
- Extended the abort scenario shell command from a 5-second sleep to 60 seconds.
- Raised the corresponding test timeout to 15,000 ms for the abort test case.
- Changed hashline format to canonical `§` section headers and `"`/`"`/`≔` operations across grammar, parser, and docs.
- Reworked range and op parsing so single anchors are valid, legacy `-`/`-=` ops error, and empty `≔` payloads now delete ranges.
- Removed legacy `HL_EDIT_SEP`, `$HSEP$`, and `hsep` plumbing, adopting raw payload lines.
- Updated execution, input, renderer, and streaming flows to use `HL_FILE_PREFIX` and `HL_OP_CHARS` helpers.
- Updated session-stats parsing to `§`/`≔`/`"`/`"` format, patch envelopes, and bumped parser versions.
- Removed the hashline-separator benchmark script and its PI_HL_SEP job orchestration.
- Added a shared `interruptHint()` utility in the modes shared module to generate the interrupt suffix with themed bracket glyphs.
- Replaced hardcoded working-message interrupt text in interactive and event controllers with calls to `interruptHint()`.
- Updated working-message rendering to recognize and strip the new themed hint when applying shimmer styling.
- Updated OpenAI completions parameterization to honor `disableReasoning` on effort-based compatible models by sending the minimum supported effort.
- Expanded commit and title generation token budgets so reasoning models can return output after internal thinking while non-reasoning calls keep existing limits.
- Switched title generation to request a `set_title` tool call, added extraction from tool-call arguments, and updated tests for the new behavior.
- Updated interactive mode to render the loader spinner with the current working message accent when available.
- Added a fallback to theme accent and a reset color code when no working-message accent is provided.
- Loading message rendering now derived session-specific accent colors and applied them to shimmer output when a session name was available.
- Shimmer palettes were updated to accept raw ANSI color values as well as theme color names during compilation.
- A unit test was added to confirm shimmer text rendered with a supplied ANSI crest color.
- Added SettingsList#setItems to replace items and clamp selection to a valid index after updates.
- Updated SettingsSelector to rebuild active memory items on backend changes and skip refresh when appropriate.
- Switched MCP wizard and command spinners to theme frames with themed initial frame and 80ms updates.
- Reworked welcome intro animation for a 3-second eased sweep with optional shine blending.
- Added memory backend refresh tests and aligned package changelogs with the updated behavior.
- Added optional `onBeforeYield` configuration and `setOnBeforeYield` in Agent, executed before follow-up checks.
- Added `YieldQueue` to `AgentSession`, with setup/teardown and streaming/idle flush via `setOnBeforeYield`.
- Replaced immediate async-result follow-up dispatch with queued batch entries, including stale-state suppression.
- Added MCP follow-up queueing in SDK, deduplicating updates by `serverName` and `uri`.
- Added changelog entries for `onBeforeYield`, async-result batching, MCP dedupe, and `display.shimmer` modes.
- Added yield queue unit tests for streaming emission, debounced idle batches, stale filtering, and error isolation.
- Added `display.shimmer` setting (`classic`, `kitt`, `disabled`) with default `classic` and UI metadata.
- Added shimmer mode resolution with defaulting plus classic/KITT intensity tier profiles and thresholds.
- Added `getFgAnsi` handling across shimmer compilation and progress-bar themes, with fallback ANSI output defaults.
- Reworked shimmer rendering to coalesce same-tier segments, cache palettes by symbol, and map disabled mode to mid-tier.
- Tuned loader timing to a 16ms render interval and 80ms spinner stepping for smoother ~60fps updates.
can1357