#requestAcpPlanApprovalChoice returned 'value !== REFINE_OPTION' which
treats any non-refine outcome — explicit dismissal, transport failure,
abort, or timeout — as approval. That meant closing the elicitation
dialog (or the request failing) granted the agent write access without
explicit user consent.
Tighten to 'value === APPROVE_OPTION' so only the explicit approve
selection passes the gate. Refine, dismissal, and every other outcome
fall through to refine semantics: the caller keeps plan mode active and
returns guidance text instead of renaming the plan and exiting.
Adds a regression test that drives a form-capable harness with a
cancelled elicitation and asserts the plan file is preserved, plan
mode/handler stay active, plan reference stays unset, and no mode-exit
notifications are emitted.
- Removed the AssistantMessageComponent completion flag and its stable-line counting helpers.
- Added a TUI regression test validating that reflowed streaming rows leaving the viewport are preserved in native scrollback and remain reachable.
- Removed changelog notes for obsolete stable-prefix scrollback APIs.
- Inserted guidance to plan only request-critical work and defer changelog, test, and documentation tasks.
- Added a dedicated cleanup phase requiring tests, docs, changelogs, and scaffold removal after smoke-test verification.
- Ensured cleanup work is treated as mandatory finalization rather than part of upfront planning.
- Marked steering user messages and wrapped them pre-LLM so the model sees a `` envelope.
- Kept transcripts and persisted history with the user's original raw text.
- Restored `AssistantMessageComponent` stable-prefix completion API.
ACP plan approval exited plan mode internally and emitted the current-mode
notification, but it did not push the matching mode config-option update.
Clients that cache or render the config selector could therefore keep
showing mode=plan after approval switched the session back to default.
The approval path now emits the same config_option_update used by other
mode transitions, and the regression test asserts that the approval-driven
exit updates both current_mode_update and the mode config option.
ACP plan mode set the plan-mode state but never installed the standing
resolve handler that interactive mode wires in #enterPlanMode. The agent
in plan mode dutifully called resolve { action: 'apply', extra.title }
to submit its plan; ResolveTool.execute then consulted
peekQueueInvoker() ?? peekStandingResolveHandler(), found neither, and
threw 'No pending action to resolve. Nothing to apply or discard.' —
stranding the agent with no path out of plan mode in Zed (and any other
ACP client).
#applyModeChange now registers a standing handler when entering
mode: 'plan' and clears it when leaving. The handler:
- validates the plan file exists at the configured '/data/workspaces/can1357__oh-my-pi__1869/.omp-session/2026-06-04T15-12-15-302Z_019e9331-31c6-7000-9bce-6be119505a1d/local' path,
- normalizes the agent-supplied title to a safe filename stem,
- asks the ACP client to confirm via unstable_createElicitation when
the client advertises elicitation.form (auto-approves otherwise so
the agent never gets stuck on a client without the surface),
- renames the plan to '/data/workspaces/can1357__oh-my-pi__1869/.omp-session/2026-06-04T15-12-15-302Z_019e9331-31c6-7000-9bce-6be119505a1d/local/<title>.md,'
- sets the plan reference path so the next turn injects the plan as
context, then clears the standing handler + plan-mode state and
emits current_mode_update so the client UI reflects the exit.
Refinement (user picks 'Refine plan') returns guidance text and leaves
plan mode active so the agent can iterate.
Fixes#1869
- Registered and triggered a session-switch reconciler during init and after switch.
- Rebuilt resume flow to restore plan/goal mode state and clear stale mode flags.
- Marked fallback and context-promotion model switches ephemeral and skipped them on restore.
- Added regression tests for temporary model ordering, fallback precedence, and mode cleanup.
- Mapped left/right arrows to switch tabs in the extension dashboard.
- Updated agent and extension dashboard footers to show arrow controls.
- Reformatted HTTP transport assignment and reordered an import.
- Aligned handoff, reminder, and system-prompt expectations with shortened copy.
- Added HTTP transport test for required initialize failures.
- Guarded SSE startup timeout against stale connection races.
- Added system-prompt rule mapping user `parallel`/`parallelize` requests to `task` subagent fan-out.
- Clarified that parallel tool calls alone do not satisfy the request.
- Mapped the right arrow key to advance tabs in the agent dashboard.
- Mapped the left arrow key to reverse tabs alongside shift+tab.
- Kept existing tab and shift+tab handling intact for backward-compatible navigation.
- Adjusted session and dashboard renderers to derive heights from live terminal rows.
- Propagated terminal-row callbacks through picker/controller wiring and session selectors.
- Reworked list visibility and page navigation to honor row-based budgets and footer lines.
- Added DECRQM 2031 startup probing and stopped OSC11 polling once support was confirmed.
- Raised default search results to 20 and context size to high.
- Added related questions parsing and return_related_questions request flag.
- Added API-key request-shape tests covering defaults and parsing.
- Added token-aware query matching so highlights mirror `HistoryStorage.search` tokenization.
- Added token highlighting, compact relative timestamps, and selected-row styling in results.
- Added PageUp/PageDown and Home/End selection shortcuts in `HistorySearchComponent`.
- Added selector matchers for `tui.select.pageUp` and `tui.select.pageDown`.
- Changed empty-result messaging to distinguish `No history yet` from `No matching history`.
- Added tests for page/home/end navigation and token-highlighting result rendering.
- Replaced the single-line description Input with Editor in agent-dashboard.ts for multiline entry.
- Added Enter-newline and Ctrl+Enter submit handling to the create-agent editor flow.
- Updated agent-creation-architect.md guidance for concise single-sentence whenToUse output.
- Added integration tests for multiline editor rendering and Ctrl+Enter submission.
- Added assistant-message completion state and stable-line count reporting.
- Added transcript stable-row aggregation and snapshot replay to keep unstable rows out of scrollback.
- Added TUI stable scrollback boundary APIs and stable-prefix commit path.
- Added regression tests proving unstable rows stay out of scrollback until stabilized.
Route '/data/workspaces/can1357__oh-my-pi__1863/.omp-session/2026-06-04T13-36-11-717Z_019e92d9-3fc5-7000-a66f-15cad94b7e75/local' plan artifacts through OMP's session-local storage instead of the editor writeTextFile bridge, preserving ACP bridge routing for regular editor-visible files.
Fixes#1863
- Removed early FTS-only return so prefix and substring matches merge.
- Sorted merged results globally by prompt recency before limiting.
- Added /force, /copy, and /shake tips.
Reviewer noted that an unconditional ExaProvider.isAvailable steered the auto chain into the public MCP fallback before any later-configured provider could run. Restored the credential-gated isAvailable, then split out isExplicitlyAvailable so resolveProviderChain still routes an explicit Exa selection through MCP without affecting other providers.\n\nRefs #1860
- 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.
Restored Exa's unauthenticated MCP fallback when neither auth storage nor EXA_API_KEY provides credentials. Preserved API-key search ordering and bounded the MCP request with the web-search hard timeout. Updated provider copy and regression coverage for the no-key path.\n\nFixes #1860
- Converted compaction, branch, and handoff prompts to fragment voice.
- Replaced "You MUST" phrasing with bare "MUST" directives.
- Applied same rewrite to autoresearch and turn-aborted prompts.
MCP tool renderers now merge call and result components so completed blocks replace the pending call header with success or error status.
MCP protocol errors now preserve top-level isError so downstream render state can use the error status consistently.
Added renderer coverage for completed MCP success and error blocks.
Fixes#1855
Centralized append-only context auto-mode resolution so SDK sessions, interactive sessions, and the status display use the same provider checks. Auto mode now enables for Xiaomi/SGLang hosts and explicit stored-request compat signals while preserving DeepSeek behavior.
Added focused regression coverage for Xiaomi Token Plan SGLang HiCache endpoints and explicit on/off behavior.
Fixes#1851
Allowed OpenAI-compatible providers to opt into Anthropic cache_control markers via compat.cacheControlFormat while preserving OpenRouter Anthropic defaults.
Fixes#1845
Marketplace plugin toggles in Settings → Plugins wrote the installed-plugin
registry and then only requested a render through the settings callback. In a
live TUI session that left already-loaded slash commands, hooks, tools, agents,
and SSH tool state unchanged until the user manually reloaded plugins or
restarted.
Make the plugin-change callback awaitable and wire the settings selector to the
same reload pipeline used by the slash-command runtime: clear plugin-root caches
for the active project registry, refresh slash command state, refresh the SSH
tool, then request a render. Await the callback after npm and marketplace
registry writes so Settings → Plugins reflects live plugin state immediately.
Add a regression test proving a marketplace toggle calls setPluginEnabled before
the plugin-change reload callback resolves.
`MarketplacePluginDetailComponent` rendered the raw `installPath` from
the installed-plugin registry, which leaks the user's home directory
into the TUI when the plugin lives under `~/.omp/cache`. Route it
through `shortenPath` so $HOME becomes `~` before the value reaches
`Text`, matching the repo's TUI sanitization rules (per review on #1844).
`PluginSettingsComponent.#showPluginList()` only inspected the npm plugin
registry, so the TUI's `Settings → Plugins` panel rendered "No plugins
installed" whenever the user had no npm plugins — even when many
marketplace plugins were installed. The `/plugins list` slash command,
`omp plugin list` CLI, and the install/uninstall selectors already
merged `PluginManager.list()` with `MarketplaceManager.listInstalledPlugins()`;
the settings panel was the only surface that missed the marketplace
registry when it was added.
Extend the panel to consume both data sources:
- Replace `PluginListComponent`'s `InstalledPlugin[]` input with a tagged
union `PluginListEntry = { kind: "npm"; plugin: InstalledPlugin } | { kind:
"marketplace"; plugin: InstalledPluginSummary }`. Each row now carries an
`[npm]`/`[marketplace]` kind badge, a scope tag, and the shadow indicator
for user installs overridden by a project entry. Widen the SelectList
primary column so long `name@marketplace` ids stay readable.
- Add `MarketplacePluginDetailComponent`: single `Enabled` toggle plus the
read-only `InstalledPluginEntry` metadata (version, install path,
installed-at, last-updated, git commit SHA). The toggle calls
`MarketplaceManager.setPluginEnabled(pluginId, enabled, scope)` — same
path `/plugins enable|disable` uses.
- Update `PluginSettingsComponent` to await both `PluginManager.list()` and
`MarketplaceManager.listInstalledPlugins()` in parallel, build the merged
list, and route selection to the kind-specific detail view. The empty
state now lists both install commands (npm and marketplace).
Fixes#1842
The beam backend never invoked the embedding pipeline during normal
operation: `remember()`/`rememberBatch()`/`updateWorking()` skipped `embed()`
entirely and `recall()`/`recallEnhanced()` never called `embedQuery()` on
the query text. As a result `memory_embeddings` stayed empty in every
deployment and recall silently degraded to FTS-only regardless of the
configured provider (fastembed, OpenAI-compatible API, custom).
- Added `scheduleEmbedding` on `beam.pendingExtractions` (mirroring
`scheduleFactExtraction`) and wired it from `remember`, `rememberBatch`,
`updateWorking`, and `consolidateToEpisodic`. Writes
`INSERT OR REPLACE INTO memory_embeddings(memory_id, embedding_json, model)`
with the active runtime-options model, captured before the AsyncLocalStorage
scope exits and re-entered inside the task.
- Auto-derived `queryEmbedding` inside `recall()` via `embedQuery(query)` when
the caller did not pass one. `queryEmbedding: null` is preserved as the
explicit FTS-only opt-out; `undefined` triggers auto-derive.
- Propagated `queryEmbedding` through `Mnemopi`'s `toRecallOptions` so the
facade no longer strips the override on the way to the beam layer.
- Made `Mnemopi.recall`/`recallEnhanced`/`search`/`query`, the module-level
exports, `BeamMemory.recall`/`recallEnhanced`, the free `recall`/`recallEnhanced`,
and `orchestrateRecall` async. MCP `handleToolCall`/`callToolJson`/`handleJsonRpc`
follow suit so the recall handler can await.
- Fixed `withBeam`/`withSharedBeam` to defer `beam.close()` until the async
handler resolves; otherwise the new async recall hit
`RangeError: Cannot use a closed database`.
- Updated CLI, MCP entrypoints, coding-agent `MnemopiSessionState`, and every
affected test to await the new shapes.
Verified with a new regression suite (`test/issue-1832-embedding-population.test.ts`)
exercising both ends of the bug: empty `memory_embeddings` and zero
`dense_score`.
Fixes#1832
can1357