- Added a unified eval framework with parser grammar, backend interfaces, and JS/Python execution result types.
- Added eval tool docs and updated prompts for fenced cells, `eval.py`/`eval.js`, and fallback behavior.
- Replaced the built-in `python` tool with `eval` across registry, rendering, interactive modes, and tool settings.
- Migrated Python execution runtime from `src/ipy` to `src/eval/py`, renamed state fields, and removed legacy introspection.
- Refactored browser tooling from in-process VM helpers to worker-managed tab supervisors and protocol transport.
- Added eval parser fallback and JS tool-bridge tests, updated imports, and removed obsolete python-mode suites.
- Added a localProtocolOptions field to session and executor option types for configurable local:// behavior.
- Passed localProtocolOptions through to LocalProtocolHandler creation and subagent session bootstrap.
- Propagated parent session local protocol settings from TaskTool so subprocess subagents share the same local:// artifacts and session context.
- Updated `formatMatchLine` to emit `*` for matched lines, a leading space for context, and a `|` anchor/content separator.
- Revised grep/hashline mismatch messages and prompts to describe the new marker and separator format.
- Aligned affected atom and hashline tests with the updated match-line prefixes and separators.
- Added `AgentRegistry` singleton with session registration/unregistration and IRC routing metadata for peer lookups.
- Added IRC messaging prompts and tooling with `irc.enabled` setting, `list/send` tool paths, and peer roster rendering.
- Changed `/btw` to session-side `runEphemeralTurn`, added background IRC exchange flushing, and fixed empty-input checks.
- Added unit tests for IRC tool and BtwController ephemeral behavior, including disabled, busy, not-found, and abort cases.
- Renamed subagent completion flow from `submit_result` to `yield` across SDK tools, prompts, and docs.
- Updated executor/task handling to require and parse `yield` calls, replacing legacy submit-result extraction and state flags.
- Added `subagent-yield-reminder` and updated system prompts to require `yield` with `result.data` or `result.error`.
- Renamed hidden-tool and registration plumbing to `yield`, including discovery helpers and renderer/test surface.
Subagents created with enableMCP=false had toolSession.mcpManager
unset, causing depth-2+ sub-subagents to re-discover and spawn
duplicate MCP server processes.
- Add mcpManager option to CreateAgentSessionOptions
- Set toolSession.mcpManager unconditionally after MCP block
- Pass options.mcpManager from executor to createAgentSession
- Guard callback registration: only wire onToolsChanged/onPromptsChanged/
onResourcesChanged when the session owns the manager (created via
discovery), not when reusing a parent's — prevents child sessions
from clobbering the parent's live MCP refresh handlers
Latent since 91da560cc (in-process subagent migration), observable
since f82a5d121 added task.maxRecursionDepth allowing depth-2 agents.
- Standardized missing-file read errors and now return `File not found: <path>` for absent edit targets.
- Centralized AI provider, usage, and OAuth helpers into shared modules to remove duplicated logic.
- Migrated OAuth/API-key login flows to shared factory helpers and removed inline prompt/token-exchange code.
- Reused shared tools and formatter utilities for discovery, stream tails, LSP batching, and source formatting.
- Consolidated repeated test helpers and fixtures into shared modules, replacing inline helper duplicates.
- Removed `SearchDb` APIs and `searchDb` fields, dropping db-backed state from native and agent sessions.
- Replaced crate export `fff` with `fd`, moving fuzzy-find bindings into `fd.rs`.
- Removed `SearchDb`/picker fast-path logic from `glob` and `grep`, simplifying scan flow and dropping db args.
- Removed `SearchDb`/`getSearchDb` wiring from extension, tool, and task context constructors across coding-agent.
- Added over-indentation validation warnings in chunk-edit normalization for suspicious `~` body line formatting.
- Removed `bytes`, `fff-grep`, `fff-search`, and `blake3` deps, adding `grep-searcher = "0.1"`.
- Passed the user context when setting the session name during subprocess execution.
- Updated native build scripts to import detectHostAvx2Support from the correct shared module paths.
- Added bash.autoBackground.enabled and bash.autoBackground.thresholdMs settings with defaults for background job behavior.
- Added auto-backgrounding support for long foreground bash commands with managed job state and timeout threshold.
- Updated bash prompts, job-protocol messages, and tool activation checks to use async and auto-background support.
- Added background bash completion integration with AsyncJobManager and end-to-end tests for short/long auto-background scenarios.
- Document session name getter/setter methods in extensions.md
- Add stub methods to ExtensionProxy that throw if called before init
- Add delegating implementations to ExtensionProxyInit
- Wire up getSessionName and setSessionName in extension runtime
- Add method signatures to ExtensionContext interface and type
- Add getSessionName to session manager API
- Add getSessionName and setSessionName to all mode contexts:
- ACP agent
- Extension UI controller (also updates terminal title)
- Print mode
- RPC mode
- Extracted prompt rendering and formatting utilities from coding-agent to centralized pi-utils package with new API surface (prompt.render, prompt.format, prompt.registerHelper).
- Migrated parseFrontmatter utility from coding-agent to pi-utils package; updated 8 files to import from @oh-my-pi/pi-utils.
- Removed 170-line prompt-format.ts module and consolidated 192 lines of Handlebars helper registrations into pi-utils prompt module.
- Updated 60+ files across coding-agent and typescript-edit-benchmark to use new prompt.render() and prompt.format() API from pi-utils.
- Simplified prompt-templates.ts by delegating core functionality to pi-utils while retaining custom helper registrations (jtdToTypeScript, jsonStringify, etc.).
- 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)
- Added `wait_for_picker_scan()` function to search_db module with cancellation token support for polling picker scan completion.
- Integrated picker-based file search into glob matching logic with `collect_files_from_picker()` helper to reuse shared SearchDb picker results.
- Refactored fff and grep modules to use centralized `wait_for_picker_scan()` wrapper instead of direct FilePicker calls, improving cancellation handling.
- Propagated SearchDb instance through agent session initialization and input controller to enable unified file picker coordination across search operations.
- Added `attribution` option to `PromptOptions` for controlling billing and initiator attribution.
- Updated subagent prompts to explicitly set `attribution: "agent"` for accurate billing attribution.
- Refactored message attribution logic to use configurable `promptAttribution` with fallback defaults.
- Added test coverage for `attribution` option in subagent reminder prompts.
Fixes#439.
feat(coding-agent): added attribution option and explicit session directory control
- Added `attribution` option to `PromptOptions` for explicit billing and initiator attribution control.
- Updated `SessionManager.create()` to require both `cwd` and `sessionDir` parameters for explicit session directory control.
- Changed session directory naming for temporary working directories from `--` format to `-tmp-` prefix.
- Made `cwd` and `sessionDir` fields mutable in SessionManager to support session relocation.
- Fixed automatic migration of legacy session directories to new `-tmp-` prefixed naming scheme.
- Updated all test fixtures to pass both `cwd` and `sessionDir` parameters to `SessionManager.create()`.
- Fixed boolean type coercion in fetch and executor modules by wrapping truncation flags with Boolean() cast.
- Removed maxBytes property from truncation metadata to simplify output metadata structure.
- Normalized optional result properties with explicit fallbacks in output-meta module.
- Updated test expectations to reflect undefined truncation properties instead of false/null values.
- Added optional `assignment` field to task result and progress interfaces to track raw per-task assignment text separately from full templated task.
- Updated task rendering to display assignment text instead of full task template when available, improving clarity of task display.
- Modified task section rendering to show trimmed assignment text with fallback to task field if assignment is not available.
- Propagated assignment field through task executor, template renderer, and result objects to maintain consistency across task processing pipeline.
- Added 'todo.eager' configuration setting to automatically create a comprehensive todo list after the first user message.
- Added 'buildNamedToolChoice' utility function to build provider-aware tool choice constraints for named tools.
- Modified tool choice resolution to support per-turn tool choice overrides via consumeNextToolChoiceOverride() method.
- Implemented eager todo enforcement mechanism that injects a synthetic prompt to encourage todo creation when conditions are met.
- Extracted tool choice building logic into reusable utility module for better code organization.
- Added comprehensive test coverage for eager todo enforcement functionality in AgentSession.
- Introduced Effort enum and ThinkingConfig metadata for per-model reasoning capabilities with min/max effort levels.
- Migrated thinking level API from string-based ThinkingLevel to structured Effort enum across agent and AI packages.
- Added model-thinking module with effort mapping, policy application, and semantic versioning utilities for provider-specific thinking modes.
- Removed supportsXhigh() function and replaced effort clamping with model-aware validation using ThinkingConfig metadata.
- Expanded models.json with thinking configuration objects for 50+ models including Claude, Gemini, and OpenAI variants.
- Added Python analysis scripts for edit tool usage patterns and tool invocation stream processing.
- Updated option interfaces, param builders, and stream functions to use unified `reasoning` field.
- Added `resolveOpenAiReasoningEffort()` to centralize xhigh clamping logic.
- Replaced type casts with `castApi()` helper and fixed test option names.
- Updated coding-agent and benchmark references for consistency.
- Extracted thinking module with ThinkingEffort, ThinkingLevel, and ThinkingMode types to centralize reasoning configuration across packages.
- Migrated ThinkingLevel type from pi-agent-core to pi-ai package with new validation functions parseThinkingLevel() and getAvailableThinkingLevel().
- Consolidated thinking level constants and descriptions into reusable exports (ALL_THINKING_LEVELS, THINKING_MODE_DESCRIPTIONS) for consistent UI display.
- Removed local thinking-effort-label utility and replaced formatThinkingEffortLabel() with centralized formatThinking() function from pi-ai.
- Refactored thinking mode handling to distinguish ThinkingSelector (user-facing with 'off' option) from ThinkingEffort (provider-level).
- Add dereferenceJsonSchema() that inlines local $ref pointers and strips
$defs/definitions from MCP tool schemas before they reach LLM providers.
Previously, Anthropic's convertTools() extracted only properties/required,
dropping $defs and leaving dangling $ref — the LLM never saw the actual
type definitions (e.g. SourceAnchorInput enum values from nucleus).
- Silence Ajv logger (logger: false) on all three instances that use
strict: false. MCP servers may declare non-standard format keywords
(e.g. "uint") that caused console.warn() to corrupt TUI output.
- Cache compiled Ajv validators per schema object identity in validation.ts,
eliminating redundant recompilation on every tool call.
Co-authored-by: Miroslav Drbal <miroslav.drbal@gendigital.com>
- Add a dedicated abortReason field to SingleResult and thread it through task execution paths so aborted subagents show actionable context instead of a generic badge.
- Populate abortReason for signal cancellation, pre-start cancellation, submit_result aborted status, and missing submit_result after reminders
Fixes#248
- Added public `waitForIdle()` and `getLastAssistantMessage()` APIs to AgentSession for deterministic session state access.
- Refactored deferred continuation scheduling from raw `setTimeout()` to centralized post-prompt task tracking system for concurrent recovery operations.
- Fixed race conditions between deferred TTSR/context-promotion continuations and `prompt()` completion via shared recovery orchestrator.
- Replaced `#waitForRetry()` with `#waitForPostPromptRecovery()` to unify retry and TTSR resume gate handling.
- Removed per-task skill pinning feature; subagents now inherit session skill set instead of per-task selection.
- Removed `preloadedSkills` option from `CreateAgentSessionOptions` interface and related system prompt plumbing.
- Removed `skills` field from Task schema; task execution now passes available session skills directly to subagents.
- Simplified task execution pipeline by removing skill resolution logic and preloaded skills handling from system prompt templates.
- Added async background job execution for bash and task tools with configurable concurrency limits and automatic result delivery.
- Added cancel_job tool and /jobs slash command to manage and inspect running background jobs with status display.
- Added jobs:// internal protocol handler for querying job status and retrieving job execution details.
- Added async.enabled and async.maxJobs settings to control background job execution behavior.
- Enhanced status line to display count of running background jobs with visual indicator.
- Implemented AsyncJobManager with exponential backoff retry delivery, job lifecycle tracking, and automatic eviction.
Fixes#56.
- Fixed submit_result tool to only terminate on successful execution instead of always terminating.
- Removed deferred termination logic and simplified abort behavior to call requestAbort immediately.
- Added submitResultCalled flag tracking to properly manage tool execution state.
- Added test coverage for submit_result tool retry behavior after execution errors.
- Exported `finalizeSubprocessOutput()` function and `SubmitResultItem` interface for subprocess output finalization with submit_result validation.
- Added automatic reminders (up to 3) when subagent stops without calling submit_result tool, aborting with exit code 1 after final reminder.
- Extracted subprocess output finalization logic into dedicated `finalizeSubprocessOutput()` function for improved testability and reusability.
- Added comprehensive test coverage for subagent warning injection, reminder behavior, and abort handling in executor module.
- Added validation to submit_result tool to ensure status field is present and correctly typed as 'success' or 'aborted'.
- Fixed executor to only mark submitResultCalled when submit_result tool succeeds or aborts, preventing false positives on validation failures.
- Added type guards and error state checks to prevent setting completion flags on malformed tool execution results.
- Added comprehensive test coverage for submit_result extraction with valid and malformed payload validation.
- Added MCPRequestOptions interface with signal property for request cancellation via AbortSignal.
- Added abort signal support to MCP tool execution enabling request cancellation via Escape-to-interrupt or other abort mechanisms.
- Enhanced MCP request handling with abort signal propagation through HTTP, SSE, and stdio transports with proper cleanup.
- Improved stdio transport request handling to use Promise.withResolvers for cleaner async flow and better abort signal integration.
- Updated HTTP transport to combine operation abort signals with timeout signals using AbortSignal.any() for unified cancellation.
- Modified SSE response parsing to support abort signals and distinguish between timeout and user-initiated cancellation.
- Added `temperature` configuration setting to control LLM sampling temperature with values from 0 (deterministic) to 1 (creative) to -1 (provider default).
- Added temperature option selector in settings UI with preset values: Default, 0, 0.2, 0.5, 0.7, 1.
- Renamed `settingsInstance` parameter to `settings` in `CreateAgentSessionOptions` for consistency.
- Updated all internal references from `settingsInstance` to `settings` throughout SDK and components.
- Integrated temperature setting into agent configuration and selector controller.
- Migrated console.error() calls to structured logger.warn() and logger.error() throughout codebase.
- Updated os.tmpdir() import style in browser tool from destructured to namespace import.
- Replaced ASCII ellipsis characters (three dots '...') with Unicode ellipsis character ('...') throughout the codebase for improved typography.
- Adjusted string truncation logic to account for single-character Unicode ellipsis instead of three-character ASCII ellipsis, reducing reserved space from 3 to 1 character in truncation calculations.
- Updated truncation offsets in multiple files (session-manager, agent, executor, footer) to preserve 2 additional characters before ellipsis due to more compact Unicode representation.
- Fixed task executor to properly handle agents calling `submit_result` with null data by treating it as missing and attempting to extract output from conversation text rather than silently failing.
- Added documentation caution in task.md about schema vs agent mismatch causing null output, with guidance on using `schema` parameter to override built-in schemas.
- Added warning message when subagent calls submit_result with null data to help diagnose schema mismatch issues.
- Renamed web search types and functions to remove 'Web' prefix for broader applicability (WebSearchProvider -> SearchProviderId, WebSearchResponse -> SearchResponse, WebSearchTool -> SearchTool, etc.).
- Refactored web search provider system from object-based configuration to class-based architecture with abstract SearchProvider base class and concrete provider implementations.
- Refactored ModelRegistry to use direct constructor instantiation instead of discoverModels() helper function, simplifying model discovery pattern.
- Refactored config system to use new ConfigFile class with schema validation, caching, and multi-format support (JSON, JSONC, YAML).
- Simplified Exa API key discovery to check environment variables only, removing .env file reading logic.
- Refactored task API to use 'assignment' field instead of 'args' for per-task instructions, enabling clearer separation between shared context and task-specific work.
- Introduced structured context/assignment separation pattern with '<swarm_context>' wrapper for template rendering, replacing placeholder-based substitution.
- Changed agent frontmatter field from 'thinkingLevel' to 'thinking-level' (kebab-case) for consistency with YAML conventions.
- Removed 'context' parameter from ExecutorOptions as context is now prepended at template level rather than executor level.
- Removed 'args' field from AgentProgress and SingleResult interfaces, simplifying task result tracking.
- Updated task rendering to display full task text instead of formatted args, improving clarity in progress output.
can1357