gen:bundle previously ran gen:docs and gen:docs:reset around the build. Under prepack that reset ran before bun pm pack assembled the tarball, so the published src/ copy of docs-index.generated.txt was empty and consumers importing via subpath (e.g. @oh-my-pi/pi-coding-agent/internal-urls/docs-index) fell through to the missing repo docs/ tree.
Restored gen:docs in prepack so postpack owns the reset (matching the original lifecycle), and made gen:bundle assert the embed is populated and fresh instead of touching it, so a direct gen:bundle outside prepack fails with an actionable message.
Fixes#3934
Added a contributor-facing native crate map (docs/native-crates.md) covering pi-natives, pi-shell, pi-ast, pi-iso, pi-walker, pi_uu_grep, pi-uutils-ctx, and vendored brush crates, and linked it from natives-architecture.md and user-facing-packages.md.
Added a docs-index tool coverage test asserting every BUILTIN_TOOL_NAMES entry and injected custom tool (generate_image, tts) has a docs/tools/<name>.md page served by omp://.
Inlined tiny fail/buildPayloadText/checkDocsIndexFreshness helpers in generate-docs-index.ts per the project rule against single-expression named functions.
Fixes#3934
Added root omp docs for memory_edit, learn, manage_skill, generate_image, and tts, plus package-level coverage for user-facing README-only CLIs.
Added a docs-index freshness check to package check and made gen:bundle generate and reset the docs embed itself.
Fixes#3934
- Streamed parameter chunk deltas as `toolArgDelta` events during execution for Anthropic and DeepSeek dialects.
- Integrated streaming tool call tracking in `InbandStreamProjector` to preserve native tool identifiers and partial JSON arguments.
- Added message validation functions to detect and drop malformed tool calls with empty or whitespace-only IDs.
- Introduced comprehensive suite of integration and unit tests validating argument delta streaming and tool-call sanitization.
- Added `#streamTurnNonce` to prevent aborted streaming turns from corrupting content indexes of subsequent messages.
- Implemented temporary stream-key generation using content position and turn nonces for previewing tool calls without native IDs.
- Added migration logic to key pending tool previews by their real ID and rekey `ToolArgsRevealController` once the real ID is parsed.
- Removed the ignore-based fallback walker implementation and its associated dependencies.
- Simplified the directory traversal to consistently use native scanners across all platforms.
- Eliminated `WalkError::Unsupported` and `WalkStatus::Unsupported` error handling across all dependent crates.
- Added a depth-first sorting fallback for native traversal and implemented recursion detection using a symlink stack.
- Replaced native `AbortSignal.timeout` calls with self-clearing timeout helper functions across model discovery.
- Added `withTimeoutSignal`, `withCatalogDiscoveryTimeout`, and `withOpenAICompatibleDiscoveryTimeout` helpers to manage cancellable fetch timeouts.
- Supported `timeoutMs` options throughout the Ollama, Llama.cpp, LiteLLM, vLLM, LM Studio, and OpenAI-compatible discovery processes.
- Documented the fix addressing Bun garbage collection segfaults caused by uncancellable timeout signals.
- Integrated StressRenderScheduler into the issue #1962 test suite to control and drain render schedules deterministically.
- Updated the settle helper and TUI initialization to accept and use the mock scheduler instead of arbitrary timeouts.
- Added `scripts/**/*.ts` to the linter and formatter inclusion list in `biome.json`.
- Replaced manual array index searches with standard `indexOf` operations.
- Cleaned up unused variables, obsolete helpers, and command-line flags.
- Applied consistent formatting and code style across tooling scripts.
- Replaced standard and tokio mpsc channels with flume channels across workspace crates to simplify thread synchronization.
- Swapped standard Mutex guards for parking_lot Mutexes to avoid manual lock poisoning handling and improve performance.
- Declared workspace-wide dependencies for flume and parking_lot in root and member Cargo manifests.
- Introduced a two-pass file processing architecture with `ReadPolicy` and `FileOutcome` state tracking.
- Deferred oversized files to a second pass where only their leading segment is read and searched.
- Integrated the two-pass processing logic into both sequential and native parallel grep execution paths.
- Updated agent tool definitions and user-visible messages to reflect partial coverage of large files instead of skipping them.
- Anchors the incomplete-todo reminder block inside the scrollback transcript instead of a floating live container.
- Eliminates duplicate reminder copies piling up in terminal scrollback during terminal reflows.
- Removes the dedicated `todoReminderContainer` and simplifies state synchronization on todo reload.
- Updates tests to verify sequential reminders commit as separate blocks and are left intact when tools succeed.
- Replaced custom fast-walk and fs-cache implementations in pi-natives with a new dedicated pi-walker library.
- Integrated the thread-safe, parallel pi-walker library across pi-natives, pi-shell, pi-uu-grep, and uu-find.
- Rewrote file search, fuzzy finding, and glob-matching logic to leverage pi-walker configurations and visitor traits.
- Optimized shell process tracking in pi-shell by replacing global descendant-diff logic with an isolated, per-run SpawnRegistry.
- Added parallel rayon-based walking and optimized fast paths for directory scanning and entry classification.
Added setup.cfg and pyrightconfig.json to PYTHON_ROOT_MARKERS so pyright, basedpyright, and pylsp project shapes also probe Windows .venv/Scripts before PATH fallback.
Fixes#3916
Added Windows virtualenv Scripts directories to local LSP command resolution so project-local Ruff launchers are discovered before PATH fallback.
Fixes#3916
- Refactored `formatProgressLine` to display passes and failures in the visual style of `bun test` including elapsed time.
- Implemented `extractFailingTests` to parse individual test failures and their raw output details out of captured stdout/stderr.
- Added a `formatSummaryFooter` to output a clean concluding tally of passed/failed chunks and the total wall time.
- Added a `SpawnObserver` trait to intercept freshly spawned external command PIDs and process group IDs.
- Exposed `set_spawn_observer` and `spawn_observer` on `ExecutionParameters` to allow tracking process lifecycles.
- Triggered the observer on successful external command execution unless the process is detached or reparented.
- Introduced the `FencedThinkingScanner` class to track nested Markdown code fences inside thinking blocks.
- Integrated the new scanner into Gemini and generic inband thinking parsers to prevent premature termination of thinking blocks.
- Added comprehensive unit and integration tests to validate nested fence retention and inline close resolution during stream parsing.
- Added a helper to build the Git process environment that explicitly clears common ambient Git environment variables.
- Applied the new environment builder to both async and synchronous Git commands to prevent environment bleeding from parent processes.
- Migrated the `/resume` session selector from an inline component to a fullscreen overlay.
- Enabled alternate screen buffer borrowing and mouse tracking support for the picker.
- Ensured proper cleanup of the fullscreen overlay during normal termination or shutdown.
- Configured the selector layout to pin keybinding hints and the footer to the bottom of the screen.
- Added `local` and `local-ts` modes to `ci-test-ts.ts` to coordinate full local runs of TypeScript and Rust packages in a single progress stream.
- Implemented a quiet output formatter that displays a one-line progress entry and defers verbose output until failure.
- Included previously skipped workspace packages `packages/mnemopi` and `python/robomp/web` in the local testing scope.
- Configured colorized ANSI formatting for progress logs when running in interactive TTY terminals.
- Bound repo-level scripts and rust task orchestration command under the consolidated runner.
- Added a platform-native fast filesystem traversal implementation leveraging `getattrlistbulk` on macOS, `getdents64` and `statx` on Linux, and `NtQueryDirectoryFile` on Windows.
- Integrated the fast traversal path as a preferred, high-performance fallback before standard walk operations in both cache collection and streaming search routines.
- Consolidated grep search logic by extracting a unified file-matching helper to support both standard and fast-walk execution paths.
- Introduced platform-specific Windows system dependency configurations and unit tests validating correct hidden-file filtering during scanning.
- Raised `GEMINI_HEADER_RUNAWAY_THRESHOLD` from 10 to 24 to avoid false-positive interrupts on legitimate, complex reasoning blocks.
- Added a regression test verifying that 10 distinct, progressing headers do not trip the detector while 24 headers still trigger it.
- Introduced a model canonicalization helper to strip redundant model compatibility fields that match defaults.
- Regenerated the models catalog JSON to eliminate over eight hundred lines of redundant compatibility specifications.
- Updated the variant collapse logic to rebuild models using the projected compatibility configurations.
- Added a missing type annotation to a test environment variable to resolve a compilation warning.
- Added configurations for `anthropic/claude-sonnet-5` under openrouter, vercel-ai-gateway, and zenmux providers.
- Reduced model pricing and cost structure rates for `anthropic/claude-sonnet-5`.
- Removed `trustExplicitThinkingOnly` compatibility flag from several Claude and Gemini model entries.
- Removed legacy `disableStrictTools` property from model definitions and updated tests.
- Fixed model builder variant collapse logic to properly map `compatConfig` to `compat`.
- Sanitized environment variables in git-clone test helpers to avoid host-leakage in test runs.
- Added configuration metadata for Claude 3.7 Sonnet, Claude 3 Opus, Claude 3 Sonnet, and a Kilo-hosted Claude Sonnet 5 model.
- Updated the Anthropic provider descriptor to include environment variables and catalog discovery options.
- Added test coverage verifying Anthropic provider first-party catalog discovery options.
- Added Claude Sonnet 5 model variants to the Anthropic and Devin provider catalogs.
- Seeded Claude Sonnet 5 under `ANTHROPIC_CURATED_FALLBACK_MODELS` with adaptive thinking configurations.
- Added Gemini 3.1 Flash Lite Image model specification to the Kilo provider catalog.
- Updated pricing profiles for existing catalog models and added verification tests for the new Sonnet contract.
Extended the mid-prompt /skill:<name> parser exclusions to also defer
to the bash tool (!cmd / !!cmd) and the python tool ($ cmd / $$ cmd
followed by ASCII whitespace), so drafts like '!echo /skill:reviewer'
are no longer consumed as skill invocations before the local-execution
branches of the interactive submit path get to dispatch them.
${HOME}-style shell expansions and prose-leading $ characters (which
pythonCommandPrefixLength already declines) keep matching mid-prompt
skills as before.
Fixes#3913
Restricted mid-prompt /skill:<name> parsing to non-slash drafts so
builtin/custom slash-command arguments such as /compact /skill:foo are
not intercepted before the command dispatcher runs.
Added parser and RPC dispatch regression coverage for the precedence
case while keeping leading /skill:<name> (including leading whitespace)
working.
Fixes#3913
The mid-prompt slash skill autocomplete added in #3654 replaced the
entire editor draft with /skill:<name> on accept so the dispatcher
(which only matched leading /skill:) would still fire. That wiped
every keystroke the user had typed before reaching for the skill.
Insert the /skill:<name> token at the cursor in the TUI editor —
replacing only the partial /sk slash token, leaving prose before and
after intact — and extend the skill-command parser so a /skill:<name>
token surrounded by whitespace is recognized as an invocation too,
with the surrounding prose threaded through to the skill as args.
The parser change is shared across all three dispatch sites
(interactive TUI, ACP, RPC) via a new parseSkillInvocation helper
in extensibility/skills, so the three Map<string,string> /
session.skills lookups stay aligned on the same parse.
Fixes#3913
- Removed instructions prohibiting mocks and testing defaults in favor of using a tester agent when available.
- Streamlined delivery contract rules by removing the explicit instruction against suppressing tests.
- Updated verification claim guidelines to emphasize smoke testing.
- Added the `llama3.2:3b` model configuration pointing to the quantized `onnx-community/Llama-3.2-3B-Instruct-ONNX` repository.
- Registered the model in both the available local models registry and list of valid memory model values.
- Documented the new option as a shipped local model choice in the documentation and changelog.
- Added preprocessing to quote ambiguous plain scalars containing a colon-space sequence when standard YAML parsing fails.
- Preserves the parsed types of unaffected fields and prevents fallback warnings for common unquoted description strings.
- Added tests to verify successful recovery of unquoted values and continued fallback warning coverage for unrecoverable syntax.
- Replaced the map-based persisted message index with a Set of key identities.
- Removed the message content equality check during branch verification to avoid false out-of-order skips when display-side content variants are present.
- Skipped processing tool calls in the event controller streaming message when the tool ID is missing.
- Prevented creating orphaned empty placeholder cards caused by empty IDs during early Anthropic and OpenAI tool block streaming.
roboomp