- Removed the Loader backpressure cap so slow ConPTY paints retain the documented proportional duty cycle.
- Made WSL terminal-title working state static to avoid a second periodic OSC write loop.
Fixes#8012
- Keep terminal working titles static with a colon separator on Windows instead of scheduling animated spinner updates.
- Update terminal title builder and state machine to check the platform and bypass timer intervals on win32.
- Deduplicate terminal title writes globally across all platforms.
- Adopt `SetConsoleTitleW` via `bun:ffi` for Windows terminal updates instead of OSC writes.
Kept the bare 'π' brand per owner direction: the separator between the
brand and the session label now carries the state — '>' when it's the
user's turn (idle), animated spinner frames while working, '!' when the
agent is blocked on the user. Disabled ('tui.titleState' off) renders the
pre-state 'π: label' layout. Updated the state/runtime tests to the new
contract.
- Centralized message preprocessing for tiny models to handle noise removal, code block stripping, and context formatting.
- Updated title generation logic to support self-closing tags and improved robustness against partial markers.
- Added structured guidance and system prompts for small models to prioritize output consistency.
- Implemented a title-generation benchmark harness and expanded test coverage for message preprocessing.
- Applied the known reasoning envelope filter when deciding whether a title marker is visible.
- Added title extraction regressions for reasoning tag and reasoning fence envelopes before the visible title.
Fixes#5122
- Limited markerless fallback cleanup to leading leaked-thinking envelopes so literal reasoning syntax in plain titles survives.
- Added markerless regression coverage for think tags and thinking-fence titles.
Fixes#5122
- Parse only title markers that remain visible after leaked-thinking cleanup so markers inside leaked reasoning are skipped.
- Preserve literal reasoning tag syntax inside the chosen title and cover it with a regression test.
Fixes#5122
- Reused the leaked-thinking healer before parsing title markers so visible reasoning envelopes cannot win extraction.
- Added regression coverage for <thinking> and <think> envelopes that contain internal title tags before the real title.
Fixes#5122
Guards the run-state title fixes: an extension setTitle() override survives run-state changes and spinner ticks verbatim, is superseded by an authoritative session title, and emission dedupes. Red-verified against the pre-fix emit path.
- Replaced tool-based `set_title` invocation with XML-style `<title>` marker tags for session title discovery.
- Implemented robust JSON-unwrapping logic to handle and sanitize title generation outputs.
- Updated model registry in catalog with new model support, provider prefixes, and metadata adjustments.
- Synchronized system prompt documentation and test suites to reflect the new marker-based generation flow.
Sizing `maxTokens` off the static `model.reasoning` catalog flag cannot
distinguish a thinking model catalogued `reasoning: false` (e.g. Qwen3
served locally via llama.cpp, whose bundled jinja chat template defaults
`enable_thinking: true`) from a model that never emits thinking. The
tight non-reasoning budget was consumed by the thinking preamble before
the useful output could be emitted, so every affected call silently
failed with `stopReason: "length"`.
Drop the `model.reasoning` conditional across every affected online call
site and always reserve the reasoning-safe budget. `maxTokens` is a hard
cap, not a target — non-thinking completions still return in the tiny
happy-path budget.
Sites fixed:
- utils/title-generator.ts (30 -> 1024)
- utils/commit-message-generator (60 -> 1024)
- tts/speech-enhancer (512 -> 1536)
- auto-thinking/classifier online path (8 -> 1024); classifyLocal
keeps its separate LOCAL_ANSWER_MAX_TOKENS
- session/unexpected-stop-classifier online path (16 -> 1024);
classifyLocal keeps ANSWER_MAX_TOKENS
Fixes#4355
- Added `tiny` as a first-class model role to override online models for lightweight background tasks.
- Updated session title generation, auto-thinking difficulty classification, unexpected-stop detection, and mnemopi backend to resolve via the `tiny` role before falling back to `smol`.
- Updated configuration schema and documentation to reflect the new role precedence.
Upstream main 52b8fb156 changed normalizeGeneratedTitle to reconcile a generated title's casing against the user's message instead of forcing Title Case, and updated the tiny-text unit tests but not the title-generator/role-thinking integration suites. The rebase surfaced their stale Title-Case expectations as CI failures. Update the 6 expectations to the reconciled output; no production code change.
- Applied title-casing to `normalizeGeneratedTitle` outputs using a new internal helper.
- Adjusted tiny text and title generator tests to assert the new title-cased results.
- Added a new built-in `title` model role with `hidden` metadata and updated role definitions and schema.
- Updated title generation to resolve models in `title`, `commit`, then `smol` order and added test coverage for that precedence.
- Filtered hidden roles from selector badges and documented the new built-in role in model/settings docs.
- Added new system prompts that ask models to emit titles inside `<title>` markers when forced tool calls are unavailable.
- Updated `generateTitleOnline` to use marker-based prompting and disable required `set_title` tool calls for models that do not support forced tool choice.
- Adjusted title parsing to extract the `<title>...</title>` value and fall back to stripped marker text when wrapping tags are incomplete.
The bundled title-system.md was reworded on main (30959d3ad) after this
branch was cut, so the rebased default-prompt test asserted stale wording.
Anchor on the stable set_title instruction instead.
Addresses review feedback on #2201.
- add discovery of `TITLE_SYSTEM.md` and pass it through interactive startup context
- route custom title prompts to online and local tiny title generators via protocol
- update session-title docs and changelog with override behavior
- add tests for prompt discovery, forwarding, and fallback to bundled title prompts
Move bundled models, model cache/manager, thinking metadata, effort helpers,
provider descriptors/discovery, wire constants, and model identity utilities
into the new @oh-my-pi/pi-catalog package.
Update pi-ai to keep provider runtime/auth concerns, move catalog provider
metadata into CATALOG_PROVIDERS, and migrate coding-agent, agent, stats, docs,
and tests to import catalog values from pi-catalog.
Split coding-agent model registry helpers into discovery, roles, and models
config modules while preserving registry orchestration.
BREAKING CHANGE: @oh-my-pi/pi-ai no longer exports catalog subpaths such as
/models, /model-cache, /model-manager, /model-thinking, /effort,
/provider-models*, discovery helpers, and provider wire constants; use the
matching @oh-my-pi/pi-catalog subpaths instead.
- Routed image-gen, inspect-image, and web search providers through `withAuth`.
- Used `reuseInitialApiKey`/`createAuthStorageResolver` for force-refresh and rotate retries.
- Attached HTTP status to thrown errors so the retry classifier detects retryable failures.
Logged structured session-title generation skip and failure outcomes with session/model context, and prevented credential lookup errors from escaping into the caller's swallowed promise path.
Added regression coverage for missing and failing title credentials.
Fixes#1892
- Skipped titling for greeting/filler/empty first messages deterministically, retrying on later user messages.
- Let capable title models decline taskless input via a "none" sentinel.
- Guarded against clobbering a name set by a concurrent attempt.
- Added `stripCodeBlocks` to remove fenced code blocks before titling, preventing literal noise (e.g. version strings in UI mockups) from becoming the session title.
- Added `prepareTitleInput` composing strip and truncate steps, updated `formatTitleUserMessage` to use it.
- Added unit tests for stripping logic and an integration test verifying the model never receives code block contents.
- Added `get(path)` support in test `createSettings` to return tiny model setting for `providers.tinyModel`.
- Updated `createSettings` in tests to accept optional `tinyModel` with default `"online"`.
- Added `flushMicrotasks()` and helper factories for test settings, registry stubs, and online model mocks.
- Added tests for `raceFirstNonNull`, `generateSessionTitle` routing, and providers tinyModel enum validation.
- 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.
roboomp