- Simplified chunk path parsing to extract selector separately from CRC field.
- Refactored chunk edit operations to conditionally strip CRC for auto-accepted batch operations.
- Enhanced chunk edit preview formatting to display operation types with target and line ranges.
- Improved chunk state rendering with clamped visible range and unified notice message format.
- Refactored error handling in chunk resolution from map_err to let-else pattern.
- Added MacOSPowerAssertion native module to prevent idle-sleep on macOS during active sessions.
- Integrated power assertion lifecycle management into coding-agent session with automatic start/stop.
- Implemented N-API bindings for macOS IOKit power assertions with RAII cleanup and cross-platform no-op fallback.
- Reformatted indentation from spaces to tabs across Rust chunk module for consistency.
- Refactored chunk selector resolution to extract and preserve raw selector strings separately from parsed CRC checksums.
- Reorganized chunk state lookup infrastructure with dedicated maps for checksum, leaf, and suffix-based chunk resolution.
- Simplified batch operation CRC validation to auto-accept stale checksums for subsequent operations on the same chunk.
- Unified TypeScript interface and Nix attrset chunk classification with consistent naming and nested expression handling.
- Extracted clipboard utility functions to dedicated module with improved OSC 52 and Termux support.
- Updated chunk read formatting with anchorStyle parameter and ellipsis standardization for improved consistency.
- Removed error on stale CRC when a chunk was already modified by an earlier batch op; the model can only know the pre-batch checksum.
- `ensure_batch_operation_target_current` now returns `Ok(true)` to signal auto-accept, skipping redundant CRC validation.
- Updated prompt to clarify batched line-scoped edits are supported and to require fully qualified chunk paths.
- Replaced nested operation objects with a flat edit schema using a single `target` field.
- Merged `sel` and `crc` into a unified `target: "chunk#CRC"` string copied from read output.
- Replaced `operations` array with `edits` and added `append`, `prepend`, `after`, `before` fields for insert positioning.
- Removed `defaultSelector` and `defaultCrc` top-level params; CRC is now always explicit per-edit.
- Fixed chunk edit operations to extract and use embedded CRC checksums from selectors as fallback validation.
- Refactored patch operations to parse selectors and strip CRC suffix before processing insert, replace, and delete commands.
- Extended parseChunkSelector to return optional CRC field extracted from selector suffix for validation.
- Improved operator precedence clarity in read tool by adding parentheses around nullish coalescing expression.
- Migrated native bindings from TypeScript wrappers to NAPI-RS generated modules with auto-generated type definitions and runtime enums.
- Replaced chunk tree API with stateful ChunkState class supporting render, edit, and resolve operations with improved error handling.
- Converted callback signatures to error-first pattern (error, result) for shell, PTY, glob, and grep operations.
- Introduced type-safe enums for MacOSAppearance, GrepOutputMode, KeyEventType, ImageFormat, and AstMatchStrictness replacing string literals.
- Refactored chunk tree implementation with dedicated modules for edit, indent, resolve, and state management with comprehensive validation.
- Moved clipboard utilities from native bindings to coding-agent package with improved OSC 52 and Termux compatibility.
- Added `anchorStyle` setting ("full", "kind", "bare") to control anchor name verbosity in chunk output.
- Changed anchor tag delimiters from `` to `[…]` and gutter separator from `│` to `|`.
- Changed edit response to render only changed hunks with context instead of the full chunk tree.
- Added `read.prosechunks` setting to enable chunk rendering for prose files (markdown, text, log, asciidoc, restructuredtext) in chunk edit mode.
- Added `read.explorechunks` setting to show chunk tree without checksums for read-only agents like explore.
- Implemented prose language detection and conditional chunk skipping logic based on file type and agent capabilities.
- Fixed zero-width insert detection to derive target indent from neighboring lines instead of replaced range.
- Improved indent selection for body-level insertions by picking deeper indent between two neighbors.
- Updated error messages to clarify insert vs splice operations and line range validation.
- Modified assertChecksum() return type to handle cases with default CRC and no explicit selector.
- Distributed language-specific AST classification logic from defaults module to individual language modules for 15+ languages.
- Simplified defaults module to minimal catch-all fallbacks, removing 341 lines of match-based classification logic.
- Consolidated chunk edit operations by removing splice operation and unifying with line-scoped replace parameters.
- Simplified chunk rendering output format by reducing thresholds and changing line count display from 'lines' to 'ln' suffix.
- Removed PI_CHUNK_SPLICES environment variable and chunkSplicesEnabled conditional logic throughout codebase.
- Changed chunk edit operation format from flat `op` field to nested keyed union structure with operation name as object key.
- Added parseKeyedChunkOp() function to normalize both keyed and legacy flat operation formats for backward compatibility.
- Enhanced chunk selector sanitization with filename prefix stripping and case-insensitive CRC checksum matching.
- Updated chunk edit prompt documentation and test suite to reflect new nested operation JSON structure.
- Added `new_segment` parameter to `init_experiment` to force new segment creation when contract fields match.
- Added `skip_restore` parameter to `log_experiment` to preserve working tree state and pre-existing uncommitted changes.
- Changed `log_experiment` to revert only run-modified files instead of entire working tree, preserving user changes.
- Changed `init_experiment` to detect matching contract fields and skip re-initialization unless `new_segment=true`.
- Added git status parsing utilities to track pre-run dirty paths and distinguish tracked vs untracked modifications.
- Removed secondary metrics validation requirement, making secondary metrics informational only.
- Fixed indentation and formatting across multiple files for consistency.
- Reformatted code blocks to use tabs instead of spaces and improved line breaks for readability.
- Added connection timeout configuration to benchmark runner for early abort on no events.
- Implemented two-phase timeout strategy in benchmark runner with connection and activity phases.
- Migrated react-edit-benchmark package to typescript-edit-benchmark with pi-mono source repository.
- Added InProcessClient implementation to eliminate subprocess spawning overhead in benchmark runs.
- Extended benchmark configuration with chunk edit variant, retry limits, and conversation dump support.
- Refactored runner.ts to support both RPC and in-process client modes with improved error telemetry.
- Cleaned up 129 benchmark report files from react-edit-benchmark/runs directory.
- Added LSP diagnostic versioning support to track document versions and suppress stale diagnostics.
- Added options parameter to `waitForDiagnostics` and `getDiagnosticsForFile` with version filtering capabilities.
- Enabled `versionSupport` in LSP client capabilities to receive diagnostic version information from servers.
- Added test coverage for stale diagnostic suppression scenarios in LSP diagnostic freshness.
The agent had no way to infer the mcp:// prefix needed to read MCP
server resources. Attempting <server>://<resource> directly fails.
All other internal URL schemes (skill, rule, memory, agent, etc.) are
already listed in the Internal URLs section of the system prompt.
Added mcp://<resource-uri>. URI resolution: exact resource listings
first, then RFC 6570 URI templates advertised by connected servers.
- Fixed memory leak by cancelling idle compaction timer on event controller disposal.
- Fixed session resumption to preserve last non-empty session when starting fresh.
- Fixed stash detection to use git ref resolution instead of output parsing for reliability.
- Fixed secret obfuscation to deobfuscate restored session messages locally while keeping LLM messages obfuscated.
- Fixed stash pop operation to preserve staged changes with --index flag after task branch merges.
- Changed idle compaction settings from enum to numeric type for flexible configuration.
Ensures exit_plan_mode waits for all previous tools to complete
before checking if the plan file exists. Prevents race condition
when LLM calls exit_plan_mode before write completes.
- Fixed PR checkout tool to resolve worktree paths to canonical form using fs.realpath().
- Refactored test mocks to use vi.spyOn for git module functions instead of inline implementations.
- Updated test setup to pass Settings.isolated() with edit.manageImports enabled to EditTool.
- Added afterEach hooks across test suites to restore mocks after each test execution.
Compaction and tool-output pruning can rewrite the session JSONL in
place, shrinking the file. When stat.size < bytesRead, the cache is
now invalidated and the file is re-read from scratch, preventing the
observer from getting stuck on a stale transcript.
Only advance bytesRead to the last complete newline-terminated line,
not to EOF. A partial trailing record (read during mid-write) stays
unconsumed and gets re-read on the next refresh, preventing permanent
entry loss from the transcript.
- Move lifecycle end event after finalizeSubprocessOutput() so
submit_result({status:'aborted'}) is correctly reflected in the
observer instead of showing completed/failed (P1)
- Reset observer registry on session resume and /new command so
stale subagent sessions from prior conversations are cleared (P2)
- Cache parsed JSONL transcript incrementally: track byte offset,
only read and parse new bytes on each refresh instead of reparsing
the entire file, avoiding render loop blocking on large sessions (P2)
Wire EventBus through task runtime so subagent lifecycle and progress
events propagate to the TUI. Add SessionObserverRegistry to track
active sessions, and a session observer overlay accessible via Ctrl+S
that shows a picker of running subagents and a read-only transcript
viewer that reads the subagent's session JSONL file to display
thinking, text, tool calls, and results.
- Add TASK_SUBAGENT_LIFECYCLE_CHANNEL for start/end events
- Add SubagentProgressPayload.sessionFile for session file tracking
- Pass eventBus through sdk.ts -> main.ts -> InteractiveMode
- SessionObserverRegistry with multi-listener onChange pattern
- Overlay picker preserves selection position across live refreshes
- Viewer renders full transcript from session JSONL (thinking, text,
tool calls with smart arg summaries, inline tool results)
can1357