- Replaced WeakMap model cache with provider/id string keys for stable reuse.
- Returned official model ids directly when matched, before heuristics.
- Collapsed non-message token path to system prompt and tool schema totals.
- Shared immutable model registries and auth storage via beforeAll/afterAll.
- Swapped fixed-delay settle sleeps for predicate polling and signals.
- Stubbed network/timers to drop wall-clock waits in registry and history tests.
- Added resetDisplay invalidation tests and startup-timing breakdown lines.
- Used `||` so empty stderr falls through to abortReason in agent bridge.
- Preferred assistant errorMessage over "Cancelled by caller" on internal aborts.
- Forced `maxRuntimeMs: 0` for eval subagents via ExecutorOptions override.
- Parsed /actions/runs URLs into run and job render handlers.
- Rendered run metadata with per-job breakdown, showing steps for failed jobs.
- Fetched job logs via API token, stripping ISO timestamp prefixes.
- BLOCKING: reorder resolveProviderCredentialIdentityKey so email
identity takes priority over project — two users with different
emails on the same GCP project no longer get merged/hard-deleted.
- Added #getUsageReportScopeProjectId helper so Gemini CLI reports
(which set projectId on limit.scope but not metadata) still get
dedup coverage. Both metadata and scope projectId paths checked.
- formatAggregateAmount now falls back to limits.length when no
scope.accountId values are present, preserving pre-existing
behaviour for providers that don't set accountId on limits.
- Added 9 contract tests for the antigravity usage merge logic:
tier dedup, worst-fraction-wins, mixed-case collapsing,
reset-time-from-other-entry, windowId separation, metadata,
sort order, and null-on-no-project.
- Nits: label='Usage' (so formatLimitTitle renders 'Usage (Default)'
not bare 'Default'), id uses params.provider instead of hardcoded
string, tier field drops redundant ?? undefined.
Gemini CLI provider stores projectId on limit.scope but not in report
metadata, so the metadata-only projectId fallback added earlier missed
that case. Now all three lookup sites (dedup identifiers, TUI account
label, ACP account label) also check limit.scope.projectId.
- Antigravity usage provider now deduplicates model quota entries by tier
instead of emitting one bar per model (15+ redundant bars for one account).
The upstream API groups quota by tier — models within the same tier share
the same quota bucket, so per-model bars were misleading noise.
- Reports now carry credential email and accountId in metadata so the
/usage display and deduplicator can show meaningful account identities
instead of 'account 1'.
- formatAggregateAmount no longer uses limits.length as account count.
Instead counts unique accountId values from limit scopes — a single
account's N incomplete limits no longer display as 'N accts'.
- Usage report dedup now considers metadata.projectId for Google Cloud
providers so duplicate credential rows with the same project merge.
- account labels in both TUI and ACP markdown paths now fall back to
metadata.projectId before the generic 'account N' placeholder.
The Anthropic-compat endpoints hosted under *.xiaomimimo.com (every
Xiaomi MiMo Token Plan region plus api.xiaomimimo.com) emit thinking
blocks without a signature. convertAnthropicMessages defaulted to
"signing capable" for any endpoint not explicitly allowlisted as
non-signing, so MiMo's unsigned thinking blocks were demoted to text on
every continuation request. Without its prior reasoning replayed, MiMo
destabilized tool-call argument serialization — the root cause behind
the args?.ops?.map crash already mitigated at the renderer in #2005.
Extend isNonSigningAnthropicEndpoint to cover the xiaomi catalog
provider, every xiaomi-token-plan-* provider id, and any baseUrl on
xiaomimimo.com so the existing non-signing replay branch fires for MiMo
the same way it does for DeepSeek and Z.AI.
Fixes#2005
Python eval agent() collapsed every subagent runtime-limit abort into a
generic 'RuntimeError: bridge call __agent__ failed' instead of the real
reason. runEvalAgent built its failure message with:
result.error ?? result.stderr ?? result.abortReason ?? <default>
? is nullish-coalescing, so result.stderr = "" (the executor's value for
a runtime-limit abort) short-circuited the chain and never reached
abortReason. The host bridge then shipped {ok: false, error: ""}, and
prelude.py's '<msg> or <fallback>' picked the named-bridge fallback.
Extracted buildSubagentFailureMessage(): aborted subagents prefer the
trimmed abortReason; otherwise fall through error, stderr (trimmed),
abortReason, and the named-bridge default. Empty/whitespace strings no
longer mask anything. The failure-detection condition also accepts
result.aborted so an abort with exitCode 0 (theoretically) still flows
the abort reason out.
Added a regression test asserting that runtime-limit aborts, whitespace
stderr/error, and totally blank aborts all produce non-empty messages
matching the executor's abortReason text.
Fixes#2006
- Forced authenticated ask requests to `experimental`, matching the anonymous fallback since the cookie session ignores pro upgrades.
- Kept TUI collapsed search answers full; capping now only applies in compact mode via `maxAnswerLines`.
- Preserved full multiline task pending preview instead of bounding it.
The todo tool's renderCall ran args?.ops?.map(...) directly, which throws
TypeError on any non-array ops value. parseStreamingJson surfaces such
shapes mid-stream: a partial Anthropic input_json_delta buffer like
'{"ops":"[{' becomes { ops: '[{' }, and intermediate states can hand back
null entries before object fields arrive. Each crash spammed Tool
renderer failed warnings and starved the TUI render loop.
Guard against:
- ops being any non-array (string, object, primitive)
- entries being null / non-object
- entry.items being a non-array
The fix is in the TUI renderer only — schema validation in the agent
loop is unchanged, so any genuinely malformed model output still
surfaces an invalid-args tool error to the model.
Fixes#2005
- Sent the OAuth token as `__Secure-next-auth.session-token` cookie since the ask endpoint ignores bearer headers and silently downgrades to `turbo`.
- Fell back to `result.title` when web results omit `name`.
- Renamed `callPerplexityOAuth` to `callPerplexityAsk` and removed a stray brace.
- Added tests covering OAuth, API-key, and anonymous request shapes.
- Stopped expanded view from dumping every match when all hits share one file.
- Applied an `EXPANDED_LINES × 2` budget while keeping context rows.
- Appended a `… N more matches` summary when truncated.
- Added anonymous Perplexity authentication mode for unauthenticated web searches.
- Switched web-search setup checks to use `isExplicitlyAvailable` and removed key enforcement in doctor.
- Updated Perplexity OAuth flow to reuse auth handling for all non-key searches and anonymous responses.
- Updated CLI and provider option help text to mark the Perplexity key optional with fallback.
- Stopped prepending system_prompt to the consumer ask endpoint, which lacks a system slot and refused the meta-instruction.
- Kept system_prompt as a proper system message on the API-key path.
- Updated `install:dev` to symlink `packages/coding-agent/scripts/dev-launch` into Bun's global bin directory as `omp`.
- Added a `dev-launch` shell script that launches Bun from an isolated directory and preserves the caller's working directory for restoration.
- Added a preload shim that restores `OMP_LAUNCH_CWD` before CLI execution so external project `bunfig.toml` preloads are not used.
- Added sanitizeErrorText in render-utils to normalize and truncate tool error messages.
- Introduced formatErrorDetail for indented subordinate error text without redundant icon or Error prefix.
- Updated goal and write tool renderers to use the new detail formatter, with write now handling isError results via a status header plus detail line.
- Showed answer text in full in the TUI; kept the `omp q` compact cap.
- Rendered each source as a single title/domain/age line with the URL linked on the title.
- Collapsed the metadata block to one Provider line plus Usage.
- Rendered search errors as a framed panel matching the success layout.
- Fixed custom-rendered tools with `mergeCallAndResult` (e.g. `lsp`) emitting a redundant tool-name line above the framed result.
- Collapsed the leading blank line for self-delimiting framed boxes.
- Added gallery fidelity routing `lsp`/`task` through the custom-tool branch via a `customRendered` fixture flag.
- Added gallery harness tests guarding state coverage and the custom-branch fallback label.
- Added lazy-loaded `gallery` command registration and new filters for tool, state, width, expanded, and plain output.
- Implemented gallery state rendering with terminal-width defaults, state filtering, and unknown-tool fallback handling.
- Added shared fixture types and aggregated renderer fixtures for multiple tool families in `galleryFixtures`.
- Added tests for renderer state coverage, route-specific output (streaming/progress/success/error), and fixture fallback.
- Updated edit result rendering to inline diff change statistics in the file header instead of using a separate metadata row.
- Removed the redundant standalone metadata line and removed the extra blank line before diff bodies for a tighter single-hunk display.
- Added a test asserting the header now contains +/-/hunk stats and that no extra stats row appears before the diff.
- Added `TUI.resetDisplay()` to force an immediate full-frame replay including native scrollback.
- Moved the persistent model selector default from Ctrl+L to Alt+M, preserving existing user remaps.
- Reserved Alt+M so extensions cannot shadow the model selector shortcut.
Passed options.expanded through the edit call preview renderer so approval previews can lift the streaming diff tail window and hide the preview label.
Added regression coverage for collapsed versus expanded edit preview rendering.
Fixes#1992
- Added `setPaddingY` to `Box` and used `setBoxPaddingForFramedBlock` when rendering framed outputs.
- Read tool results now skipped vertical padding in framed blocks, removing extra blank rows above and below the output.
- Added a regression test for `ToolExecutionComponent` that confirmed framed read content no longer renders extra blank lines.
- Stopped sharing parentEvalSessionId with bridge-spawned subagents.
- Sharing it deadlocked since the parent's kernel blocks on the bridge call.
- Each subagent now gets its own eval session with an independent kernel.
- Adjusted renderCollapsedSearchGroups to compact each result group before truncation so first-section hits stay visible.
- Removed collapsed-body truncation notices and kept truncation status in the output header.
- Made PI_INTENT_TRACING, PI_AUTO_QA, PI_PY, and PI_JS take precedence when set.
- Fell back to config when the env flag is unset instead of ORing.
- Surfaced PI_PY=0/PI_JS=0 in the disabled-backend error messages.
- Emitted per-command cmd:N targets after the assistant message that issued them.
- Carried commands from text-less messages to the next visible message.
- Replaced the single trailing "last command" leaf.
- Updated EventController to keep foreground render mode active from agent_start through agent_end so live-region rebuild remains enabled during a turn.
- Pinned stop-reason assistant errors to the editor banner and suppressed the transcript's inline Error line while pinned, then restored it on the next agent_start.
- Added tests covering error-banner suppression/restoration and eager native scrollback rebuild ordering before loading animation.
- Updated write streaming content formatting to accept an expanded flag and show full output only when expansion is active.
- Passed the expanded state from the write renderer so in-flight write previews now lift the 12-line cap on Ctrl+O.
- Added streaming write preview tests to verify collapsed tail capping, expansion growth, and short-write behavior.
- Updated rawTextInputFromPartialJson to stop excluding values that start with '[' from raw text fallback detection.
- Preserved existing exclusions for inputs starting with '{' and '"' in partial JSON detection.
- Replaced `¶path#hash` prefix with `[path#hash]` delimiters across parser, tokenizer, and grammar.
- Updated prompts, docs, and recovery paths to the new bracketed form.
- Added a fullscreen /copy tree of recent assistant messages with nested code blocks and a live preview pane.
- Removed the /copy last|code|all|cmd subcommands in favor of tree selection.
- Extracted copy-target assembly into a testable util.
- Dimmed models whose context window is smaller than current usage.
- Skipped disabled entries during navigation and selection.
- Showed context-limit suffix and warning for unselectable models.
- Stopped expanding the volatile tool block so a >100-line code arg no longer overflows the viewport.
- Kept streaming and resolved render shapes consistent at codeMaxLines.
can1357