- Replaced @xterm/headless backing with libghostty-vt WASM for grapheme-aware widths.
- Removed legacy/modern width-model split now that the terminal matches the renderer.
- Reframed width-disagreement regressions around Ghostty row containment.
- Added non-destructive in-band resize handling and regression coverage.
When OTEL_EXPORTER_OTLP_ENDPOINT (or _TRACES_ENDPOINT) is set, register a
global OTLP/proto TracerProvider and switch on the agent loop's telemetry
so its invoke_agent / chat / execute_tool spans reach a collector instead
of a no-op tracer. Honors the standard OTEL_* env contract
(OTEL_SERVICE_NAME, OTEL_SDK_DISABLED, OTEL_TRACES_EXPORTER=none) and the
OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT capture toggle; no-op
when no endpoint is configured.
Uses the exporter-trace-otlp-proto 2.x line, which exports cleanly under
Bun (the 1.x line deadlocks). Adds a flushTelemetryExport() helper and a
periodic + signal-driven flush so long-lived processes surface spans.
This makes the existing telemetry usable from hosts that spawn omp as a
child process (e.g. an ACP harness), where an in-process provider
registered by the parent can't reach the child process.
- Added local Anthropic message client wrappers, request types, and retry-timeout handling.
- Replaced Anthropic SDK usage with local wire types and request schema.
- Updated AnthropicOptions.client to accept AnthropicMessagesClientLike and fallback paths.
- Removed @anthropic-ai/sdk from root and package manifests.
- Defined `EmbeddingRow` and `EmbeddingOutput` in runtime options and exported them from core embeddings.
- Updated `EmbeddingProvider`, `MnemosyneEmbeddingProvider`, and `provider` runtime option types to return `EmbeddingOutput` instead of `unknown`.
- Refactored embedding result normalization to accept typed rows and sync/async batches and coerce them into validated `Float32Array` vectors.
- Consolidated `cosineSimilarity` in `vector-math`, removed duplicate local impls, and treated mismatch/non-finite as zero.
- Switched beam, query-cache, recall, shmr, and binary-vectors to consume shared `cosineSimilarity` from `vector-math`.
- Changed embeddings to parse `$env/$flag`, return `Float32Array`, lazily import model deps, and retry via `fetchWithRetry`.
- Added `@oh-my-pi/pi-utils` and replaced hardcoded FastEmbed cache paths with `getFastembedCacheDir`.
- Expanded truthy parsing for lowercase `y`, `true`, `yes`, and `on`, then updated optional-embedding tests for cache behavior.
- Updated binary-vectors and optional-embedding tests for NaN-safe cosine, `Float32Array`, and cache-dir expectations.
- Upgraded workspace catalog dependencies to newer versions in both package manifests.
- Regenerated bun.lock to align transitive versions, including Vite and ecosystem tooling updates.
- Updated ACP startup tests and transport handling to fail fast on unsupported server transport types.
- Added tiny-title protocol contracts, including progress-state unions, message payloads, and transport interfaces.
- Added title text utilities to truncate long inputs, wrap `<user-message>` blocks, and normalize generated titles.
- Added tiny-title model registry and helpers with type-safe keys and runtime optional loading via optionalDependencies.
- Added client-side worker orchestration with spawn fallback, request queueing, progress/error routing, and smoke-test APIs.
- Added worker runtime for model resolution, prompt-based inference, lock-based install retries, and close-time cache clear.
- Added @huggingface/transformers to package.json dependencies for Transformer feature preparation.
- Updated bun.lock with new @huggingface/transformers resolution and transitive onnxruntime/sharp entries.
- Aligned lockfile package sources by converting @oh-my-pi/pi-mnemosyne from workspace to catalog resolution.
- Introduced `@oh-my-pi/pi-mnemosyne` workspace package with BeamMemory, recall/retain, FTS, and optional fastembed ONNX embeddings.
- Wired `memory.backend = "mnemosyne"` into the coding-agent settings schema and backend resolver.
- Added `MnemosyneSessionState` for per-session auto-recall on first turn and auto-retain after completed turns.
- Documented configuration, environment variables, and operational notes in `docs/mnemosyne-memory-backend.md`.
- Replaced snapshot internals with full-file records and removed contiguous/sparse snapshot APIs.
- Added file-hash normalization, computed `computeFileHash`, and updated grammar/messages to 4-hex tags.
- Simplified recovery by checking whole-file hashes first, then applying merge-replay fallback after mismatches.
- Updated coding-agent tools to use `record`/`recordFileSnapshot` and skip hash headers for unsnapshotted large files.
- Expanded patcher and snapshot tests to verify 4-hex anchors, hash deduplication, and cache-capped behavior.
can1357