- Prevents absolute `AbortSignal.timeout` from cutting off active stream bodies by introducing a clearable pre-response timer.
- Clears the watchdog timer for Bedrock, Gemini, Ollama, and Codex providers once headers are received.
- Restores regular caller abort signaling capability on the combined fetch stream.
- Adds comprehensive fake-timer tests to cover the timeout arming and clearing lifecycle.
- Extracted native token counting into a new localized `tokenizer.ts` wrapping `@oh-my-pi/pi-natives`.
- Introduced a lightning-fast byte-length estimation logic for token counting when accurate counting is disabled.
- Diverted token calculations to the faster estimator during test environments and when `PI_TOKENIZER_ACCURATE` is falsy.
- Updated agent base and coding-agent sessions to consume the new localized `countTokens` utility.
- Configured searcher line numbers based on whether the output mode is set to content.
- Updated `build_searcher` and `build_searcher_for_params` to accept and set the line number flag.
- Added detection for provider error finish reasons occurring before tool calls to identify fatal messages.
- Prevented subprocess tool execution finalization from resetting a non-zero exit code when yield items exist.
- Ensured a default error message is set in stderr when a subprocess fails after yielding a result.
- Deleted the `reasoning` object on OpenRouter completions requests when using a Kimi model with a forced tool choice.
- Prevented reasoning conflicts with Kimi by dropping `{ reasoning: { enabled: false } }` and completely omitting the parameter instead.
- Added `sessionId` and `cwd` tracking to JS evaluator session instances.
- Updated `acquireSession` in the JS context manager to update existing session identity and directory metadata on reuse.
- Forwarded the current working directory from `executePerCall` and `executeOnSession` callers to the underlying Python kernel tasks.
- Introduce a `transformAssistantMessage` lifecycle hook to `AgentOptions`, `Agent`, and `AgentLoopConfig`.
- Invoke the hook on finalized assistant messages within the agent streaming loop prior to context updates, UI emission, and tool dispatch.
- Enable mutating the text content and tool-call arguments of assistant messages in place during execution.
- Added native support for parsing, normalized validation, and serialization of ArkType schemas throughout the agent pipeline.
- Implemented pruning of unconstrained union branches and normalization helpers to handle unrepresentable strict-mode branches.
- Patched ArkType's schema package to preserve declared object key order during serialization.
- Integrated ArkType schemas into `agent-loop` and migrated coding agent tool parameters to ArkType format.
- Derived the `SEARCH_PROVIDER_ORDER`, `SEARCH_PROVIDER_PREFERENCES`, and `SEARCH_PROVIDER_LABELS` metadata dynamically from a single `SEARCH_PROVIDER_OPTIONS` source of truth.
- Reprioritized the default search provider sequence, shifting higher-order choices like Perplexity, Gemini, and Anthropic ahead of Tavily and Brave.
- Updated documentation to reflect the new search provider evaluation order.
- Added explicit ArkType schema descriptions across all coding agent tool definitions.
- Updated schema definitions in autoresearch and commit tools with descriptive wrappers.
- Documented tool schema enhancements in the packages/coding-agent CHANGELOG.
- Migrated all wire protocol, schema definitions, and tools validation from Zod to ArkType across multiple packages.
- Updated extension runtimes, custom tools loader, and TypeBox compatibility shim to expose and use ArkType instances.
- Added a comprehensive ArkType migration guide, validation parity tests, and helper utilities.
- Removed redundant PDF asset routing and parsing implementations from the read tool.
- Clarified the message instructing users or models on how to handle unseen lines.
- Added specific instructions recommending a ranged read to bypass summarization and mint a fresh tag.
This change introduces a new `openrouter` API type and extensively refactors OpenAI-family streaming providers, centralizing shared logic and improving robustness.
Key changes include:
- **Unified OpenAI-family Logic:** Consolidated core utilities, compat resolution, request shaping, and stream processing into `openai-shared.ts`, reducing duplication across `openai-completions`, `openai-responses`, and `openai-codex-responses`.
- **OpenRouter API Type:** Introduced a dedicated `openrouter` API type with dual-surface compatibility, allowing it to dispatch requests as either OpenAI Chat Completions or Responses.
- **Enhanced Provider Integration:**
- Improved Perplexity search to leverage shared OpenAI streaming transports, including API-key fallback to OpenRouter and support for Perplexity's Responses API.
- Integrated xAI-specific logic directly into the shared `stream.ts` dispatch, removing the dedicated `xai-responses` provider.
- Refined credential parsing for Google Gemini CLI and handling of Azure deployment names.
- **Robustness & Consistency:** Improved error handling for Codex, standardized output token parameter resolution, and ensured consistent application of reasoning suppression across all Chat Completions dialects.
- **New Documentation:** Added `provider-endpoint-constraints.md` to detail endpoint-specific behaviors and quirks for various providers.
- **Telemetry & Debugging:** Extended telemetry propagation to advisor calls and overflow compaction tasks. Improved debugging for Codex WebSocket failures and stream error messages.
- **Tooling & Security:** Updated browser stealth scripts to prevent detection and added a new `ts-no-inline-cast-access` TTSR rule.
- Bound native Reflect methods to local variables in the browser launch script to prevent detection.
- Updated stealth injection scripts to use the bound Reflect methods instead of global Reflect calls.
- Fixed a type assertion issue in the AgentSession tool proxy.
- Added a new builtin rule definition file, `ts-no-inline-cast-access.md`, with AST patterns that flag inline cast member access and usage guidance.
- Registered the new rule in `src/discovery/builtin-rules/index.ts` so it is included in the built-in TypeScript rule set.
- Updated `ts-no-any.md` to recommend schema parsing at trust boundaries and clarified when to choose schema parsing, guards, or unchecked casts.
- Added a top-level `ttsr` CLI command with `list` and `test` actions.
- Added snippet input handling for inline text, `--file` path, and stdin via `--file -`.
- Added test-mode context inference and result output for matched and unmatched rules.
- Added CLI tests for source inference, explicit source overrides, JSON output, and list mode.
- Wrapped MCP call and result renderers in WidthAwareText so formatting can use runtime content width.
- Computed inline argument preview budget from the available content width instead of a fixed 70-character cap.
- Updated raw text truncation in MCP results to target the measured content width while preserving expand hints and warnings.
- Introduced a WidthAwareText component that formatted fallback content during render using the computed content width.
- Updated ToolExecutionComponent to use WidthAwareText and invalidate cached text instead of a fixed-width Text path.
- Reworked generic tool output formatting to derive inline and output truncation budgets from terminal width rather than hard-coded 70/80 columns.
- Added PDF member read syntax (`doc.pdf:<member>`) and trailing-colon listing.
- Added asset-read handling to serve extracted PDF images as inline image content.
- Added PDF image extraction caching keyed by size and mtime with marker files for retries.
- Added basename validation that rejects unknown/traversal-like PDF members and shows available names.
- Added new `tui.tight` setting with default `false` and documented behavior.
- Added global tight-mode APIs and `setIgnoreTight` propagation across Box, Text, and Markdown.
- Handled `tui.tight` updates to refresh UI borders and trigger rerendering.
- Updated key message components to ignore tight mode selectively and preserve intended spacing.
Persisted the GitHub Copilot apiEndpoint when usage-driven OAuth refresh writes the refreshed credential row.
Added coverage for Business endpoint preservation through the credential-health refresh path.
Fixes#2876
- Adjusted image budget IDs to use a shared 24-bit seed and wrap safely so new budgets no longer collide.
- Added transmit-state reset and changed image output ordering so full paints can replay image data after terminal clears.
- Updated image and protocol tests to match the new wrapped-id behavior and cursor-save/restore-wrapped transmit sequences.
Stored GitHub Copilot plan-specific API endpoints discovered during OAuth login and propagated them through structured credentials, model discovery, usage/probe metadata, and request base URL resolution.
Added regression coverage for Business endpoint discovery, serialization, dynamic model discovery, and chat/responses routing.
Fixes#2876
- Added cursor save/restore escape handling around direct image emission.
- Replaced the previous up-and-down cursor movement sequence with restore logic to keep renderer output positioned correctly.
- Ensured direct image replay path no longer depends on moving back down after placement to align the cursor.
- Removed ANTIGRAVITY_NO_PREAMBLE_INSTRUCTION from the shared Gemini header constants.
- Updated Gemini CLI and web search request builders to prepend only ANTIGRAVITY_SYSTEM_INSTRUCTION before existing system parts.
- Adjusted the Gemini CLI alignment test to stop asserting the removed anti-preamble prompt lines.
- Enabled Antigravity request envelopes to include requestId/labels and per-attempt responseId state.
- Enabled budget transport to send `thinkingBudget` and capped `maxOutputTokens` 65_536 for Gemini calls.
- Added budget-aware model profiles for Gemini flash/pro variants and updated effort routing expectations.
- Refreshed variant collapse to separate Gemini and Antigravity tables and heal stale snapshots.
- Changed provider delta input comparisons in `buildResponsesDeltaInput` to use `Bun.deepEquals` instead of stringified JSON checks.
- Updated session message diffing to return early on length changes and compare normalized entries with deep equality.
- Replaced JSON-string assertions in the issue-966 repro test with `Bun.deepEquals` for stable equality checks.
- Added Gemini thinking-loop detection helpers for near-duplicate and verbatim output checks.
- Wrapped `stream`, `streamPiNative`, and `streamSimple` dispatches with the loop guard.
- Emitted retryable empty-content loop errors and stopped completion events on loop hits.
- Added `enableGeminiThinkingLoopGuard` options for OpenAI compatibility with Gemini defaults and overrides.
- Split over-limit sleep_consolidation source groups into multiple capped episodes.\n- Added regression coverage that every claimed working-memory id lands in an episode.\n\nFixes #2869
- Tracked completed primary turns in `AgentSession` and started an immune-turn window after each interrupting advisor steer.
- Routed follow-on `concern`/`blocker` notes to the aside channel while the immune window is active, while preserving prior auto-resume-suppressed handling.
- Added the `advisor.immuneTurns` setting and tests for immune-turn detection and delivery-channel decisions.
- Added Matplotlib figure PNG rendering and display tracking in Python runner to emit PNG output immediately when figures are displayed via display(fig).
- Extended session persistence to externalize oversized image payloads in both content and details.images, enabling tool result images to survive session reload.
- Enhanced session loader to resolve image data payloads and blob references across content and details.images during session reconstruction.
- Added image cache invalidation in TUI image component when image protocol, cell dimensions, or Kitty Unicode placeholder mode changes.
- Added comprehensive test coverage for Matplotlib display, image persistence across reload, and TUI image rendering with protocol and dimension changes.
- Added a configurable maxEpisodeChars cap for sleep_consolidation episodic rows.\n- Marked truncated sleep summaries in metadata before extraction, graph ingest, and embedding scheduling.\n- Covered oversized transcript consolidation with a regression test.\n\nFixes #2869
Added a shared raw-text cap for regex-only entity and pattern fact extraction so multi-megabyte transcripts skip noisy heuristic passes instead of flooding MEMORIA. Added regression coverage for the entity extractor and MEMORIA pattern fact path.\n\nFixes #2868
- Added `effortBudgets` to `ThinkingConfig` and configured Antigravity Gemini variants to use budget mode with per-effort token maps.
- Updated `mapOptionsForApi` to resolve thinking budget from request options, then model `effortBudgets`, then the provider default ladder.
- Adjusted Antigravity user-agent metadata and introduced wire profiles with IDs to support routed-model request mapping.
- Updated `executeSearch` to read `providers.antigravityEndpoint` once and catch uninitialized-settings errors, so web search fallback no longer aborts before any provider runs.
- Added missing `Vision` and `Git` entries to `TAB_GROUPS` so those new settings now render in their intended setting-section groups.
- Adjusted context and idle-compaction tests to wait for prompt-in-flight transitions and provide `getContextUsage` for deterministic snapshots during pending turns.
can1357