- Added `images.describeForTextModels` configuration defaulting to true for text models.
- Added `describeAttachedImagesForTextModel` to persist images and generate local:// descriptions.
- Added image-description notices to session flow with hidden typing and pre-user insertion.
- Added fallback behavior that returns notes when vision is unavailable or output is empty.
A semicolon-form extended color with a missing component (e.g. 38;2;255;0
lacking blue, or 38;5 lacking the palette index) has ambiguous arity: if
a following SGR is concatenated onto it, the next code is absorbed as the
missing channel/index and the rendered color changes. Stop a merge run at
any parameter list that ends mid extended-color so the following sequence
keeps its original framing. Complete colors and the self-delimiting colon
form still merge, so the win is unchanged on real transcripts (34.3% SGR
reduction on the sampled capture). Adds regression tests for incomplete
truecolor/indexed runs and the complete-color merge.
- Updated the coding-agent modes tips to mention PI_DIALECT for providers that mis-handle native tool calls.
- Clarified a troubleshooting option for restoring behavior when inference providers cannot parse tool-call payloads.
- Enabled recursive parsing in `tryParseJsonForTypes` to coerce double-encoded JSON strings.
- Updated `coerceArgsFromIssues` to parse object/array JSON strings before singleton-array fallback.
- Prevented malformed container strings from being wrapped into arrays so validation returns array errors directly.
- Added ZAI GLM-5.2 reasoning-effort mapping, translating minimal to none and xhigh to max.
- Enabled ZAI and zhipu GLM-5.2 completion requests to send reasoning_effort and tool_stream.
- Added provider token clamping so GLM-5.2 completion requests use capped max_tokens.
- Updated catalog policies to route GLM-5.2 max-token and reasoning support through ZAI/zhipu hosts.
- Removed synthetic HF model entries and aligned GLM-5.2 catalog specs with real providers.
Fixes#2833
- Changed `select_folded_spans` to skip an unfold candidate that exceeded `unfold_limit` and continue processing the remaining BFS queue instead of breaking.
- Added a regression test in `summary.rs` proving a large `<style>` raw-text span no longer starves sibling HTML sections from unfolding.
- Updated the summary unfold-ceiling setting description to document that overflowed candidates stay folded while BFS continues.
The renderer's component tree emits each styled span as
<set-color>text<reset>, producing runs of byte-adjacent SGR sequences.
Two byte-adjacent SGR sequences are semantically identical to one SGR
carrying both parameter lists (params apply left-to-right), so merging a
run into a single CSI is behavior-preserving and cuts per-frame byte
volume and SGR-dispatch count on slow terminal engines (xterm.js/WebGL
under a large viewport). Each emitted sequence is capped at 16 parameter
tokens so long runs split across several valid CSIs instead of
overflowing a terminal's parameter buffer. On by default; disable with
PI_NO_SGR_COALESCE=1.
Rewrite the local `omp stats` web UI on the OMP brand palette with a left nav, focused per-section views, a new Projects view, a system-aware light/dark toggle, and flicker-free navigation. Same data, endpoints, and command; the only server change is a 1h time-range bucketing fix in the aggregator.
- Added default system-prompt guidance to scan skill descriptions and read applicable skill:// content before work.
- Covered the rendered prompt contract with a frontend-design skill regression test.
Fixes#2829
Stopped google-gemini-cli and google-antigravity requests from sending explicit AUTO toolConfig entries for toolChoice auto, matching the shared Google provider behavior.
Added regression coverage for both Gemini CLI provider identities.
Fixes#2830
Recorded todo reminder developer messages in the session log so JSONL transcripts match model-visible context when reminders are enabled.\n\nFixes #2824
Merge displayed bridge-backed range and multi-range read lines into the existing hashline snapshot provenance so INS.POST anchors pass visible-line validation after ACP reads.
Fixes#2773
When a user who previously logged in (stored as api_key credential) re-logs
in with the new version (returns OAuthCredentials with enterpriseUrl), the old
api_key row was not replaced by #upsertOAuthCredential. AuthStorage.getApiKey()
prioritizes api_key over OAuth, so the old row was matched, returned a raw key
without JSON wrapper, and the enterpriseUrl (selected endpoint) was lost.
Changed to set() which calls replaceAuthCredentialsForProvider() to replace ALL
existing credentials for the provider.
Related to #2802
- Added a new `--advisor` flag to argument parsing and launch flag definitions.
- Applied the parsed `advisor` flag as a runtime-only override of `advisor.enabled` during startup.
- Updated advisor-related docs and added parseArgs tests for `--advisor` behavior and position handling.
- Sanitized artifact filenames by normalizing tool names before composing spill paths.
- Applied `wrapToolWithMetaNotice` to custom tool adapters and RPC-host tools in agent-session setup.
- Wrapped SDK-registered extension/custom tools with the same meta-notice adapter during session creation.
- Gave reasoning-capable local auto-thinking classifiers the safe 1024-token answer budget used by online reasoning classifiers.
- Raised the non-reasoning local classifier floor to 16 tokens and allowed tiny completions to honor larger explicit budgets.
- Added regression coverage for qwen3-1.7b and qwen2.5-1.5b local classifier budgets.
Fixes#2808
- Removed the Codex-preferred canonical remapping path for exact `provider/id` model matches.
- Resolved explicit `openai/gpt-5.5` references as `openai` provider without redirecting to `openai-codex`.
- Added regression tests to confirm explicit provider/id and enabled-model patterns are not coalesced to Codex.
- Replaced `process.chdir(tempDir)` plus `PI_REQ_DEBUG=1` auto-naming with explicit `setNextRequestDebugPath` targets, removing the `previousCwd` save/restore and the `debugFiles` directory-scan helper.
- The PI_REQ_DEBUG recording assertions now read each request/response from the path they set, so the test no longer mutates `process.cwd()` or the global debug env flag.
- Added an optional `cwd` parameter to `looksLikeLocalPath` so bare-name local-directory detection resolves against an explicit path; the install-command test now passes a temp dir instead of `process.chdir`-ing into it.
- Removed `issue-1423-repro.test.ts`, whose package-dir-lookup repro relied on `process.chdir` to stage an isolated project.
- Made `JsRuntime` track the realm globals it installs (`#globalOwner`, `#ownedGlobalKeys`, `PRELUDE_GLOBAL_KEYS`) and reclaim them in `dispose()`, re-binding ownership through `#activateGlobals()` on cwd/run-scope/run; disposing an older inline/direct runtime no longer deletes a newer runtime's helper globals, and overlapping same-realm runs are serialized via `enterGlobalRun`.
- Added `WorkerCore.dispose()` (rejects pending tool calls with `ToolError` and disposes the runtime) and invoked it from `spawnInlineWorker` teardown in `context-manager.ts`.
- Updated the `js-static-import-rewrite` test to snapshot and restore any pre-existing `__omp_import__` global instead of unconditionally deleting it.
- Added `runtime-global-dispose.test.ts` covering newer-runtime global survival, older-runtime reactivation, and cross-runtime mutation rejection.
- Clarified the `runWorkerEntrypoint` comment in `cli.ts` about `parentPort` sync-prefix buffering for tab/eval workers.
- `ProcessTerminal` now captures a per-instance `#headless` from `isTerminalHeadless()` at construction and re-reads it in `start()`; when set, `#safeWrite`, `start`, `stop`, `drainInput`, and `setProgress` short-circuit, so frame paints, `start()` probes (OSC 11 / DA1 / kitty), the progress keepalive, stdin raw mode, SIGWINCH, and teardown escapes no longer reach a real terminal during `bun test` (previously `#safeWrite` only skipped on `!isTTY`, so an interactive terminal leaked stray boxes and probe queries).
- Gated the `emergencyTerminalRestore()` blind-restore branch and `TerminalInfo.sendNotification` on the headless flag, and guarded `setTerminalTitle`/`pushTerminalTitle`/`popTerminalTitle` in `title-generator.ts`.
- Opted the terminal-contract suites (terminal-appearance, notifications, emergency-restore-altscreen, issue-2034 write gate, process-terminal-render-harness, render-stress-harness) back into real I/O with `setTerminalHeadless(false)`, restoring on teardown.
- Added `process-terminal-headless.test.ts` asserting zero writes under a forced TTY when headless and real frame/probe output after opt-out.
- Added `isTerminalHeadless()` / `setTerminalHeadless()` to `env`, a process-wide switch that suppresses real-terminal side effects (stdout escape/frame writes, stdin raw mode, CSI/OSC probes, SIGWINCH, window-title changes, emergency restore).
- Defaulted the flag to `isBunTestRuntime()` so it engages when `bun test` sets `NODE_ENV=test`, and made `setTerminalHeadless` return the previous value so callers can restore exact prior state.
Limit provider-priority ranking to provider defaults that share the first fallback default id, preserving mixed-provider startup precedence while keeping the OpenAI/Codex tie fix.\n\nFixes #2807
Route stale OpenAI GPT default roles through canonical Codex selection when the catalog prefers the Codex OAuth transport, and rank shared provider defaults by canonical provider priority.\n\nFixes #2807
can1357