The session-observer overlay is gone. The Agent Hub (ctrl+s, alt+a, or double-tap left arrow on an empty editor) presents one overlay with two views: a live registry table (status, unread irc count, current task, last activity; j/k to navigate, r to revive, x to abort/release) and per-agent chat (transcript + input line) — submitting revives a parked agent and steers it via the normal prompt path. Main is the ambient chat and stays out of the table.\n\nrenderInitialMessages no longer takes a prebuilt context: every redraw now reaches for AgentSession.buildTranscriptSessionContext() (full-history transcript with each compaction emitted inline at the point it fired, snapcompact frames re-attached on rebuild). UiHelpers drops the deferred-compaction render and the IRC autoreply branch that the new mailbox bus deprecated. CompactionSummaryMessage renders as a slim divider (── 📷 compacted · ctrl+o ──), expanding to the summary + snapcompact frame count. session-manager.buildSessionContext gains a { transcript: true } mode; an exported buildSessionContextFromFile() reads a session file without taking the writer lock so the hub chat view can tail any agent (parked or live). Theme picks up icon.camera + tool.irc symbol entries.
Address review notes on #2248:
- readTextFromClipboard's WSL branch now mirrors readImageViaPowerShell
(Bun.spawn + kill timer sharing POWERSHELL_TIMEOUT_MS) instead of
execSync, so a cold powershell.exe start cannot block the TUI event
loop on the first smart-paste miss.
- The smart-paste text fallback routes through ui.getFocused() +
hasPasteText like the enhanced-paste text path, so the payload lands
in focused modal Input prompts instead of the hidden main editor
(#2127 contract). Covered by a new routing test.
app.clipboard.pasteImage dead-ended with a status message when the
clipboard held text, which made the chord useless on hosts that only
deliver that one keypress (VS Code's integrated terminal forwarding
Ctrl+V, Windows clipboard history via Win+V simulating Ctrl+V).
- handleImagePaste now pastes clipboard text through the editor's
paste semantics when no image is available; an empty clipboard
reports 'Clipboard is empty'.
- readTextFromClipboard reaches the Windows clipboard through host
PowerShell under WSL (mirroring readImageFromClipboard), with CRLF
normalization and UTF-8 output encoding.
- handleClipboardTextRawPaste's empty-clipboard status message moved
to the actual empty branch (was shown after successful pastes).
- Clipboard reads are constructor-injectable for tests.
Implements proposal 1 of #1628.
- 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.
- Removed terminal risk mode toggles from config, terminal state collection, and render controllers.
- Dropped snapshot freezing, thaw tracking, and finalized-block replay in transcript rendering.
- Removed clear-on-shrink settings and initialization hooks from selector and interactive mode flows.
- Simplified render scheduling by using requestRender() without mutation flags or stream checkpoints.
- Updated bracketed-image parsing to recognize multiple image paths in a single paste, including quoted values and shell-escaped spaces.
- Changed the editor image-path handler to process each matched path in order, awaiting async handlers.
- Added tests for multi-path routing/normalization and recorded the fix in the coding-agent changelog entry.
On kitty (Linux/Wayland) and any terminal supporting OSC 5522 enhanced
paste, the /login API-key prompt for OpenCode Zen — and any other modal
Input prompt (Perplexity OTP, GitHub Enterprise URL, manual OAuth
redirect URL, …) — silently dropped pasted content. The user could not
paste their API key; on Enter or Esc the key surfaced in the main
prompt.
Cause: InputController.#setupEnhancedPaste enables kitty's enhanced
clipboard protocol on TUI start and consumes the resulting OSC 5522
packets in an addInputListener that runs *before* focus dispatch in
tui.#handleInput. The controller's pasteText callback then routed
unconditionally to this.ctx.editor.pasteText(text) — the main
CustomEditor — even when selector-controller had cleared editorContainer
and focused a temporary Input. The text accumulated in the detached
editor and only resurfaced when the modal was dismissed.
Fix:
- TUI.getFocused() exposes the currently focused component.
- Input.pasteText(text) mirrors Editor.pasteText so any Input can absorb
a payload from a non-bracketed transport.
- InputController's enhanced-paste callback consults getFocused() and
routes text to the focused component when it exposes pasteText,
falling back to the editor only when no modal target is in focus.
- Image pastes refuse with a status message when a modal Input is
focused, instead of stuffing a binary blob into the hidden editor.
Regression tests in packages/tui/test/input.test.ts and
packages/coding-agent/test/issue-2127-repro.test.ts pin both the new
TUI/Input contract and the InputController routing source.
Fixes#2127
- Added atomicTokenPattern support to treat image and paste markers as indivisible tokens, enabling atomic deletion on backspace and forward-delete instead of character-by-character removal.
- Extended image marker format to include pixel dimensions [Image #N, WxH] and updated paste marker format to [Paste #N, +X lines] or [Paste #N, Y chars] with consistent comma-separated metadata.
- Implemented accent styling for paste markers in editor rendering while preserving hyperlinks for image markers, improving visual distinction between marker types.
- Refactored image reference handling to unified placeholder system supporting both image and paste markers with kind-aware rendering callbacks and regex patterns.
- Passed pending and input images into compaction queueing for steering and follow-up flows.
- Delivered queued follow-up/steer prompts with stored images via session.prompt and followUp.
- Normalized empty image arrays to undefined and cleared stale pending image state.
- Added regression tests for image-backed compaction queueing and follow-up forwarding.
- Removed `<turn-aborted>` guidance injection from `transformMessages`, deleted `turn-aborted-guidance.md`, and dropped the synthetic abort note path for aborted/error turns.
- Updated `c`/`.` continue shortcuts to submit `manual-continue.md` as a hidden synthetic `developer` message via `session.prompt(..., { synthetic: true })` instead of sending an empty user turn.
- Updated tests and changelog notes in AI and coding-agent to reflect the revised abort-context and continue behavior.
- Added an interrupt state in EventController to switch the working label to `Interrupting...` and suspend intent-driven updates until the turn resets.
- Called `notifyInterrupting()` from streaming-interrupt paths in InputController and exposed it on InteractiveModeContext so Esc acknowledgement shows immediately while tools tear down.
- Added tests to verify loader acknowledgement, freeze of late intent updates during interruption, and label updates resuming on the next agent start.
Route follow-up shortcut submissions through the builtin slash-command dispatcher before queueing them as deferred prompts.
Added regression coverage for /goal set submitted through InputController.handleFollowUp while a stream is active.
Fixes#2038
handleCtrlZ called process.kill(0, "SIGTSTP") unconditionally; on
Windows the runtime rejects the signal name with TypeError, which
propagated out of the TUI input dispatcher and crashed the agent
with an [Uncaught Exception]. Even on POSIX the call could fail
(sandboxes that block sending to pid=0, runtime-reserved signals),
and a thrown error after we already stopped the TUI and registered
a one-shot SIGCONT listener left the UI stranded with a leaked
handler that would fire on the next unrelated continue and try to
re-start() an already-running TUI.
Branch on process.platform first so Windows shows a status notice
and never calls process.kill, and wrap the POSIX kill in try/catch
that removes the SIGCONT listener and re-starts the TUI on failure.
Fixes#2036
- Updated InputController streaming submit handling to interrupt and resume when queued steering exists, refreshing the pending-message UI and render.
- Added AgentSession.interruptAndFlushQueuedMessages to abort active work and continue processing queued messages immediately.
- Added tests for queued steering interruption in both agent-session concurrency and input-controller keybinding flows.
- Collapsed non-touched phases to one-line summaries in multi-phase todo renders while retaining full output for active, touched, or expanded views.
- Computed touched phases from in-progress tasks, completion transitions, and tool operations, with init operations now marking all phases.
- Added renderer tests covering collapsed rendering, argument-less fallback behavior, expanded mode, and separator-free phase output.
- Added OSC5522 parsing and a payload controller that prefers image MIME and reassembles chunks.
- Added bracketed image-path detection and `onPasteImagePath` callback wiring in the editor.
- Added local image-path paste insertion with fallback to text paste on image-size errors.
- Added `Editor.pasteText()` and OSC5522 disable sequencing on terminal cleanup.
- Added optional `reason` parameters to `Agent.abort` and `AgentSession.abort` APIs.
- Passed abort reasons through interrupt flows into underlying agent cancellation.
- Replaced hard-coded abort text with `resolveAbortLabel` for streaming and replayed messages.
- Fell back to generic `Request was aborted` text when no abort reason was provided.
- Removed the built-in `background` (`bg`) slash command and its `handleBackgroundCommand` path from the interactive flow.
- Deleted background event subscription and shutdown handling by removing `handleBackgroundEvent` from the event, input, and interactive controllers.
- Simplified `InteractiveModeContext` by dropping background-only fields and helpers such as `isBackgrounded`, background UI context creation, and background event callbacks.
- Added `TUI.resetDisplay()` to force an immediate full-frame replay including native scrollback.
- Moved the persistent model selector default from Ctrl+L to Alt+M, preserving existing user remaps.
- Reserved Alt+M so extensions cannot shadow the model selector shortcut.
- Added image-reference rendering to make `[Image #N]` placeholders clickable in chat.
- Added MIME-aware image blob materialization with extensioned sidecar paths.
- Added clickable path, line, and URL hyperlinks for read, search, and fetch outputs.
- Hardened OSC8 hyperlink emission with URI validation and control-byte/idempotency checks.
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.
When the autocomplete popup is visible, ESC unconditionally falls through
to the editor base class so it can dismiss the popup. Only an ESC with no
popup visible reaches onEscape and routes to the global interrupt handler.
Removed the shouldBypassAutocompleteOnEscape callback that paved over the
popup-dismissal path whenever the agent was busy (streaming, bash, /btw,
auto-compaction, etc.) — that is precisely the moment the user is most
likely to hit ESC while the popup is open, and dismissing-then-aborting in
one keystroke is a footgun, not a feature. Two-press semantics now match
the standard TUI/IDE pattern: first ESC closes the popup, second ESC
aborts.
Tests cover both branches in custom-editor-keybindings.test.ts and the
input-controller escape suite no longer asserts the dead callback.
Fixes#1655
- Add new keybinding `app.clipboard.pasteTextRaw` with Ctrl+Shift+V / Alt+Shift+V defaults
- Implement `readTextFromClipboard()` utility supporting macOS, Windows, Linux (Wayland/X11), and Termux
- Integrate raw paste handler into CustomEditor and InputController
- Text is inserted without formatting collapse to preserve whitespace and newlines
- Updated toggleToolOutputExpansion to pass allowUnknownViewportMutation when requesting render.
- Added a test that verifies tool output toggling sets expansion state and calls requestRender with the new flag.
- Documented the Ctrl+O POSIX tool-result expansion scrollback fix in the changelog.
- Dropped `onShowHotkeys` callback and its binding from `CustomEditor` and `InputController`.
- `?` now inserts a literal question mark regardless of editor state; use `/hotkeys` explicitly.
- Added regression test confirming `?` is treated as plain input when the editor is empty.
- Restricted `setModel` to persist settings only when `persist: true` is passed; all runtime switches (Ctrl+P, `--model`, `/model`, model picker temp selections) no longer overwrite `modelRoles.default`.
- Changed `cycleRoleModels` to accept a direction ("forward"/"backward") instead of a `temporary` flag; both directions now use `applyRoleModel` without persisting.
- Added `persist: true` exclusively to the model picker's "Set as default" action in `SelectorController`.
- Added test suite covering persistence behavior for `setModel`, `cycleRoleModels`, and `cycleModel`.
- Added `/omfg <complaint>` command integration from slash registry to mode context and input handling.
- Added OMFG panel and controller for live draft streaming, up to three retries, and confirmed save flow.
- Added strict OMFG rule extraction and validation with JSON parsing, alias checks, and history-based repair.
- Fixed auto-thinking restore to preserve resolved effort instead of reverting to pending auto sessions.
- Added a shared `renderSegmentTrack` helper to render colored segment tracks with a filled active chip style.
- Updated hook selector and role-cycle status-line rendering to use the shared track renderer and aligned role-cycle output.
- Added contrast text-color logic to `Theme` so active chip labels stay legible across fill colors.
- 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 `providerRetryWait` and `retryWait` hooks to stream/usage options so tests bypass real scheduler delays.
- Parameterized GitHub Copilot poll intervals and Copilot model retry base delay for fast test execution.
- Replaced `Bun.sleep`/`setTimeout` polling loops with `AbortSignal` event listeners in agent session tests.
- Consolidated auth-gateway E2E helpers into a shared `test/helpers` module, eliminating duplicated `checkGatewayAvailable` implementations.
- Migrated credential-disabled tests from SQLite-backed stores to an in-memory store, removing temp-dir lifecycle overhead.
- Updated the TUI shutdown slash command handler to return a `SlashCommandResult` instead of `void`.
- Returned `commandConsumed()` after clearing the editor and invoking runtime shutdown.
- Imported `SkillPromptDetails` as a type in the input controller message imports.
- Adds omp acp subcommand that launches the agent as an ACP stdio server
- Registers the subcommand in the CLI dispatcher
- Threads terminal-auth args and ACP flags through the launch and main orchestrators
- Exports AgentSession on the public SDK surface
- Updates skills loader to support skill→slash-command conversion and prompt injection
- Updates input-controller to dispatch ACP built-in slash commands
Makes `/skill:<name> [args]` work identically under both submission
keybindings, mirroring how free text is already routed during streaming:
- `/skill:foo` + Enter, streaming -> steer queue (interrupt)
- `/skill:foo` + Ctrl+Enter, streaming -> followUp queue
- `/skill:foo` + Enter, idle -> idle prompt
- `/skill:foo` + Ctrl+Enter, idle -> idle prompt (was: literal text)
A single private helper `#invokeSkillCommand(text, streamingBehavior)`
on `InputController` handles the dispatch; the Enter submit handler
calls it with "steer", and `handleFollowUp` calls it with "followUp"
after the compaction short-circuit so a skill typed during compaction
rides the same `queueCompactionMessage` queue as free text.
Behavior deltas vs upstream/main:
- Enter on `/skill:foo` during streaming now steers (was: queued as
followUp). Users who relied on the followUp default can press
Ctrl+Enter -- the same key they already use for free-text follow-ups.
- Ctrl+Enter on `/skill:foo` is new capability; previously the
literal string `/skill:foo ...` was sent as plain followUp text and
the skill was never invoked.
Op: extend
Anthropic counts sessions by metadata.user_id. Without this fix, OMP
generated fresh random entropy on every API request, inflating the
session count and preventing backend attribution to the authenticated
account.
Changes:
packages/ai:
- resolveAnthropicMetadataUserId() now accepts JSON-format user_id
matching real Claude Code's getAPIMetadata shape
({ session_id, account_uuid, ... }). Previously only the legacy
cloaking format was accepted on OAuth, causing stable caller-supplied
values to be silently discarded.
- AnthropicOAuthFlow.exchangeToken() and refreshAnthropicToken() now
populate OAuthCredentials.{accountId, email} from the token response
account block, removing the need for a separate /api/oauth/profile
round-trip.
- AuthStorage.getOAuthAccountId(provider, sessionId) returns the OAuth
accountId for the session-sticky credential, used to build
account_uuid in metadata.user_id. Guards against misattribution for
API-key, runtime-override, env-key, and fallback-resolver paths that
do not record a session credential.
packages/agent:
- Agent.metadataForProvider(provider) resolves request metadata for
the given provider via the installed resolver, or returns the static
metadata value. The plain metadata getter now returns only the static
value; provider-aware resolution is explicit.
- Agent.setMetadataResolver(fn) installs a (provider: string) resolver
evaluated per LLM request in agent-loop, after getApiKey records the
session-sticky credential, so account_uuid reflects the credential
actually used.
- AgentLoopConfig.metadataResolver is called with config.model.provider
after getApiKey, overriding the static metadata field.
packages/coding-agent:
- AgentSession.#syncAgentSessionId installs a metadata resolver that
builds { user_id: JSON.stringify({ session_id, account_uuid? }) },
matching the Anthropic session attribution format. account_uuid is
only included for provider="anthropic" to avoid leaking the OAuth
identity to third-party Anthropic-format-compatible providers.
- sessionId getter prefers providerSessionId when supplied via
AgentSessionConfig so all API paths (getApiKey, direct calls,
metadata resolver) share the same provider-facing session ID.
- prepareSimpleStreamOptions stamps session metadata on direct calls
(runEphemeralTurn, compaction, branch summary, title generation) so
they share the same session bucket as Agent.prompt requests.
- generateBranchSummary and generateSessionTitle accept a
(provider: string) metadata resolver evaluated after their own
getApiKey call for correct credential attribution.
- Added hideThinkingSummary options across stream, agent, and session payload paths.
- Routed Coding-Agent hideThinkingBlock toggles to agent hideThinkingSummary during session updates.
- Updated OpenAI, Azure OpenAI, and Codex requests to omit reasoning.summary when hide/ summary is null.
- Reworked system-prompt preparation with per-step timeouts, fallback defaults, and step-level warnings.
- Added session-draft persistence methods in SessionManager to write unsent editor text to an artifacts-sidecar draft file and delete it after single-shot consumption.
- Persisted editor text during interactive shutdown and restored that draft on resume when the editor was empty, enabling Ctrl+D draft recovery.
- Updated Ctrl+D handling in the editor/controller path and added tests covering draft round-trip, artifact cleanup, stale-draft eviction, and in-memory no-op behavior.
- Added `recordLocalSubmission` and `withLocalSubmission` to replace inline signature set mutations across input and compaction flush paths.
- Signatures are now cleaned up automatically on delivery failure, preventing stale entries from suppressing editor-draft protection on retries.
- Extended test fixtures and added cases covering signature lifecycle for idle, streaming, and fire-and-forget submission paths.
- Removed title-source aware branching from session terminal-title and accent helpers, and updated callers to use session name plus cwd only.
- Dropped UUID-based recent-session naming by preferring explicit header titles or first user prompts and generating an "Untitled · <time>" fallback.
- Adjusted welcome session-row rendering for width-aware name truncation and disabled reasoning in title generation requests to keep terminal titles concise.
- 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.
- Updated loop command handling to toggle `/loop` without a prompt argument and prompt for the next user input to start repeating.
- Stored each user-submitted prompt as the active loop prompt while loop mode is enabled so iterations auto-resubmit that input after each yield.
- Introduced `pauseLoop` behavior to clear the captured loop prompt and cancel pending auto-submit when Escape is pressed, while `handleLoopCommand` now no longer disables mode automatically with arguments.
- Tracked locally submitted user signatures for both optimistic and streamed-queue submissions in interactive-mode context state.
- Updated user message_start handling to avoid re-clearing the editor and re-adding chat for locally originated messages.
- Cleared consumed local signatures on queued-message restore and added tests for queued, external, and optimistic message_start editor behavior.
can1357