- Fixed help rendering so `--help` no longer triggers unrelated command loaders.
- Fixed startup span logging to emit markers only with PI_DEBUG_STARTUP set.
- Fixed logger startup trace behavior for `:start`, `:done`, and `:fail` phases.
- Fixed prompt template processing with cached raw-template compilation and safer formatting.
- Optimized symbol and tag parsing in prompt templates via manual parsers.
Stopped task-agent discovery from loading direct .claude/agents roots, while preserving OMP-native .omp agents and Claude marketplace plugin agents. Added a focused regression test for user and project Claude Code custom agent files.\n\nFixes #2209
Same shape of bug the reviewer flagged for custom tools: forwarding
`LoadExtensionsResult` from parent to subagent reused Extension instances
whose factories closed over the parent's `ExtensionAPI` — cwd, eventBus,
and runtime all pointed at the parent. Any tool/handler/command that
referenced `api.exec()`, `api.events`, or `api.runtime` still acted on the
parent session/worktree from inside an isolated subagent.
Forward only the path list; each session rebuilds extensions through
`loadExtensions` so factories see the right `ExtensionAPI`.
- `extensibility/extensions/loader.ts`: extract `discoverExtensionPaths`
(FS scan only) from `discoverAndLoadExtensions`. The combined helper now
composes the two. New export added to the package barrel.
- `sdk.ts`:
- Add `discoverSessionExtensionPaths()` (the `disableExtensionDiscovery`-aware
path-only counterpart of `loadSessionExtensions`).
- Add `preloadedExtensionPaths?: string[]` to `CreateAgentSessionOptions`.
Three loader branches: `preloadedExtensions` (CLI same-process reuse,
still shallow-cloned), `preloadedExtensionPaths` (subagent: skip scan,
reload locally), or full discovery.
- Document `preloadedExtensions` as same-process-only; subagent
forwarding MUST use `preloadedExtensionPaths`.
- `tools/index.ts`: `ToolSession.extensionsResult` → `extensionPaths:
string[]` for the same reason.
- `task/executor.ts` and `task/index.ts`: forward `extensionPaths`. Drop
the forward for the isolated `runSubprocess` branch — worktree cwd ≠
parent cwd, so the subagent re-discovers extensions against its own
tree.
- New `test/sdk-extensions-per-session-binding.test.ts` pins the contract:
two `loadExtensions` calls on the same path with different `cwd` and
different `EventBus` instances yield distinct Extension + runtime
objects whose factories close over the per-call bindings.
- Updated `executor-pass-through` and `sdk-preloaded-extensions-isolation`
tests for the new option name and comment context.
Refs PR review on #2193
Reviewer flagged that forwarding `LoadedCustomTool[]` from a parent session
to a subagent reused tool instances whose factories had closed over the
parent's `CustomToolAPI` — `cwd`, `exec`, `pushPendingAction`, and `ui` all
pointed at the parent. In isolated tasks the tool would `exec` against the
parent worktree and queue pending actions on the parent session.
Forward only the path list; let each session rebuild tools through
`loadCustomTools` so factories see the right `CustomToolAPI`.
- `extensibility/custom-tools/loader.ts`: extract `discoverCustomToolPaths`
(FS scan only) from `discoverAndLoadCustomTools`; export
`ToolPathWithSource`. The combined helper is now `discoverCustomToolPaths`
+ `loadCustomTools`.
- `sdk.ts`: replace `preloadedCustomTools` (`LoadedCustomTool[]`) with
`preloadedCustomToolPaths` (`ToolPathWithSource[]`). The custom-tools
block runs `loadCustomTools` unconditionally; only the path scan is
skipped when the caller pre-discovered it.
- `tools/index.ts`: `ToolSession.loadedCustomTools` →
`ToolSession.customToolPaths` for the same reason.
- `task/executor.ts` and `task/index.ts`: forward `customToolPaths`.
Drop the forward for isolated subagents — the worktree shifts `cwd`, so
the subagent re-discovers tools against its own working tree.
- New `test/sdk-custom-tools-per-session-binding.test.ts` pins the contract:
two `loadCustomTools` calls on the same path with different `cwd` and
different `pushPendingAction` callbacks yield distinct tool instances
whose factories see the per-call bindings.
- Updated `executor-pass-through` and `sdk-preloaded-extensions-isolation`
tests for the new option name and added a `ToolPathWithSource` fixture.
Refs PR review on #2193
Each `runSubprocess` call re-ran `loadCapability<Rule>()`,
`loadSessionExtensions()`, and `discoverAndLoadCustomTools()` because
`ExecutorOptions` and the `createAgentSession()` call inside the executor
omitted three pass-through fields the parent had already paid for. The
already-correct paths (skills, context files, workspace tree, MCP manager)
showed the intended pattern.
- Cache `rules`, `extensionsResult`, and `loadedCustomTools` on the
parent's `ToolSession`.
- Add `rules` / `preloadedExtensions` / `preloadedCustomTools` to
`ExecutorOptions`; forward them from both `runSubprocess` call sites
in `task/index.ts` and into the executor's `createAgentSession()`.
- Add `preloadedCustomTools` to `CreateAgentSessionOptions` and skip
`discoverAndLoadCustomTools()` when it is supplied.
- Shallow-clone `extensionsResult.extensions` when reusing
`preloadedExtensions`, so the per-session autoresearch + custom-tools
inline wrappers never leak back into the caller's array.
Fixes#2190
- Updated `renderResult` to inspect `isError` on detail-less task results and render an error state with the error glyph.
- Replaced success/completed status symbols in agent, review, and result outputs with the shared `status.done` symbol.
- Added tests for detail-less task rendering to verify failed results show error and successful results show the done glyph.
- Added status.done and tool.* symbols to theme mappings and presets.
- Replaced generic success glyphs with contextual +/-, tool icons, and warnings.
- Mapped tool/task/job completions to status.done or status.enabled with icon overrides.
- Triggered runtime provider refresh after extension registration and warned on failure.
- Updated `convertToLlm` logic to emit `developer` role for custom, hook, and file-mention inputs.
- Simplified OpenAI compact output filtering to retain only `user` and `assistant` messages, removing legacy `system-reminder` pattern checks.
- Adjusted compaction and session tests to match the new developer-role mapping and expected compacted content.
- Propagated `schemaOverridden` from `YieldTool` into executor `YieldItem` metadata.
- Bypassed schema validation on override or schema-builder errors and kept payload output with success exit.
- Emitted `SUBAGENT_WARNING_SCHEMA_OVERRIDDEN` so accepted override results no longer surface as `schema_violation`.
- Reworked running-task rendering so shimmer animation is applied to descriptions instead of IDs.
- Added accent coloring for the separator and description text to keep the status line formatting consistent.
`finalizeSubprocessOutput` always spliced collected `report_finding`
entries onto a top-level `findings` array regardless of the active output
schema. A caller-supplied schema with `additionalProperties: false` and
no `findings` property would accept the raw payload in-tool (via the
`yield` validator, which only sees the pre-injection data) but then fail
post-mortem validation — emitting `schema_violation: findings: must not
be present` and propagating as a fatal `RuntimeError` through
`agent-bridge.ts` and the eval Python/JS preludes, collapsing the entire
workflow cell along with any prior successful subagent work.
`normalizeCompleteData` now takes the resolved validator and only
performs the injection when the augmented candidate validates. When the
schema rejects it, the raw payload is returned instead — which the in-
tool yield validator already accepted, so the lockstep guarantee
documented at the top of `output-schema-validator.ts` is honored.
Findings remain visible via the agent progress stream and JSONL
artifact, so no information is dropped when injection is suppressed.
Both finalize call paths (yield-success and no-yield fallback) now share
the single validator build instead of constructing it twice, and the
yield-path schema_violation branch is now reached only via the
explicit malformed-schema check, never via spurious findings rejection.
Fixes#2070
- Reduced shimmer-driven UI animations and borders to a 30fps cadence.
- Collapsed consecutive read tool calls into one finalized transcript block.
- Tracked visible assistant content to finalize and reset read groups correctly.
- Updated running task rows to static bullet markers with shimmer-only subagent labels.
- Updated task call and result rendering to process shared context with the Markdown renderer, so context sections are now displayed with proper Markdown formatting.
- Stopped shimmer animation on pending bash/eval/task blocks once async state is `running`, preventing the committed frame from freezing a transient dark border segment.
- Adjusted rule path display to fall back to a root-relative path when cwd-relative resolution is unavailable.
- Extended `OutputBlock` section metadata with an optional `separator` flag and included it in cache hashing.
- Updated output block rendering to emit an unlabeled section divider when the separator flag is enabled.
- Marked task and result render paths to request separators for non-context content sections to improve output clarity.
- Added `/tan` slash command registration and interactive handling.
- Added TanCommandController validation and async task scheduling for `/tan` dispatch.
- Added session cloning that suppresses breadcrumbs, copies artifacts, and handles abort cleanup.
- Added `promptCacheKey` support in Agent and inherited `providerPromptCacheKey` in session creation.
- Rendered shared `# Goal`/`# Constraints` brief in result frame via buildContextSection helper.
- Undid per-field double-encoding so brief reads cleanly from raw result args.
- Removed header spinner since per-agent body lines already animate.
- Wrapped run summary in bracket glyphs to match bash tool footer.
- Enabled call/result merging by setting mergeCallAndResult on TaskTool.
- Reworked task item lines into bullet lists and removed tree-style prefixes.
- Rendered task calls as framed blocks with isolated headers and pending status metadata.
- Changed result previews to hide task titles and `Tasks` headings when a result is present.
- Used `||` so empty stderr falls through to abortReason in agent bridge.
- Preferred assistant errorMessage over "Cancelled by caller" on internal aborts.
- Forced `maxRuntimeMs: 0` for eval subagents via ExecutorOptions override.
renderAgentResult and the live-progress sibling cast
extractedToolData?.yield to Array<{ data }> and called ?.map without
checking the actual runtime shape. Optional chaining only short-circuits
on null/undefined, so any stray non-array value (a single yield object
landing in the slot) made .map undefined and threw
TypeError: completeData?.map is not a function — taking down every
`review` task render.
Both sites now route through a new normalizeYieldData helper (next to
normalizeReportFindings) that returns an array of yield records: it
preserves arrays unchanged, wraps a single object as a 1-element array
so the verdict still renders, and drops primitives. Added a regression
test exercising the result branch, the progress branch, the primitive
fall-through, and the canonical array shape — all of the failing-branch
ones reproduce the crash on the pre-fix renderer.
Fixes#1987
- Skipped count/concurrency normalization when --bench is set.
- Errored when no OAuth accounts resolve for the provider.
- Updated flag docs to run one request per OAuth account.
Per PR review on #1926: a secondary in-process top-level createAgentSession() that exposes bash/task/job tools would still call AsyncJobManager.instance() at execute time, register on the primary's manager, and have the primary's onJobComplete enqueue results into the primary's yieldQueue — corrupting the owning session's conversation.
ToolSession now carries an asyncJobManager reference scoped to its session: the constructed manager for top-level sessions, the inherited singleton for subagents (so their bash/task completions still flow into the spawning conversation as before), and undefined for secondary in-process top-level sessions that found a singleton already installed. bash, task, and job tools resolve the manager through ToolSession instead of the process-global singleton, so a secondary session whose tools attempt async work fails fast with the standard "Async job manager unavailable" error instead of contaminating the primary.
When `async.enabled` is true but `AsyncJobManager.instance()` returns
`undefined` (orphaned-session state, host that never wired one up, etc.),
the `task` tool was returning a hard error and was unusable for the rest
of the session — even though the existing sync codepath (`#executeSync`,
which still parallelizes via `mapWithConcurrencyLimit`) was right there.
Fall back to `#executeSync` instead and emit a `logger.warn` so the
missing-manager state stays diagnosable. Background/job-poll semantics
are lost in this degraded mode, but the tool keeps working.
Fixes#1922
- Added `loadOverallPlanReference` to resolve a session plan reference from local storage and skip empty or missing files.
- Updated task execution to read the active plan reference (except in plan mode) and pass it into each spawned subagent.
- Extended the subagent system prompt and session SDK/tools plumbing so subagents receive and render the approved plan path and contents.
- Added `isReadOnlyAgent` and `READ_ONLY_TOOL_NAMES` to classify agents.
- Marked read-only agents and forbade edits, commands, and reasoning offload.
- Added tests for capability classification and description rendering.
- Renamed `TodoWriteTool` to `TodoTool` and its source/prompt files.
- Updated tool registration, schema, renderers, and gating to `todo`.
- Adjusted cursor provider native tool names and tests to match.
- Renamed strike-animation constants and `todo-error-reminder` type.
- Changed `AgentOutputManager` to use requested names verbatim, adding `-2`/`-3` suffixes only on repeats (e.g. `Anna`, `Anna-2`).
- Renamed main agent id from `0-Main` to `Main`; nested ids now use dot notation without numeric prefix (e.g. `Parent.Child`).
- Updated task widget to render dotted hierarchy as `Parent>Child` breadcrumb without leading index.
- Resume scan now tracks seen names instead of a counter to avoid clobbering prior outputs.
The catch around the subagent yield-reminder prompt previously logged
every exception at ERROR. User cancel (^C) and compaction-driven aborts
both surface as ToolAbortError through awaitAbortable, so benign control
flow generated 9 spurious 'Subagent prompt failed' errors in 2 days on
the reporter's instance.
Gate the ERROR branch on '!abortSignal.aborted && !(err instanceof
ToolAbortError)' and route the abort path to logger.debug. The outer
catch + finally still mark the run aborted, so observable behaviour is
unchanged.
Fixes#1623
- Added `repairDoubleEncodedJsonString` to unescape fields double-encoded by the model (e.g. literal `\n`, `\"`, `\uXXXX` in `context`/`assignment`/`description`).
- Scoped repair to natural-language fields only, leaving code-bearing tools untouched.
- Applied repair on both render and execution paths in `TaskTool`.
- Converted [SECTION]...[/SECTION] markers to "SECTION\n===" format in system prompt templates.
- Updated system conventions doc to reference the new marker style.
- Updated tests to match against the new header pattern.
- Stored mcpManager and localProtocolOptions on ToolSession so nested subagents inherit them without relying on process-global singletons.
- TaskTool now uses the session's localProtocolOptions and mcpManager when spawning sub-tasks, falling back to defaults if absent.
- Added orchestrate keyword detection and notice handling for non-synthetic prompts.
- Added orchestrate notice handling in session output paths, including streaming and append delivery.
- Added a system orchestrate notice specifying task-subagent delegation, phase workflow, and validation gates.
- Added shared gradient-highlighter utilities and switched ultrathink highlighting to use cached palettes.
- Removed embedded orchestrate prompt artifacts and updated usage tips for orchestration, ultrathink, and /login behavior.
- Added parentMnemosyneSessionState propagation from session state through SDK, executor, and task options into nested sessions.
- Added getMnemosyneSessionState() and rekeying logic to refresh Mnemosyne IDs during session sync, switch, and restore.
- Added Mnemosyne reset and teardown cleanup on unaliasing or restoration to avoid stale state.
- Propagated task `renderContext` through `ToolExecutionComponent` so call rendering can detect result state.
- Suppressed task call-preview rows when a result snapshot exists to avoid duplicate task lines.
- Added formatContextUsage to render context as `%/window` with `?` fallback across status and task views.
- Updated task renderCall to show dispatched agents as tree entries with `Tasks (2)` header and `#3` fallback.
- Capped task preview collapse at 12 entries and added `... N more agents` overflow messaging.
- Replaced hard-coded dot separators with `theme.sep.dot` in subagent cost and context output.
- Added tests for streaming task preview rendering and updated nested-live expectations for percent/context output.
- Added `mnemosyne.scoping` setting: `global`, `per-project`, and `per-project-tagged`.
- `per-project-tagged` writes to a project-local bank while merging global memories on recall.
- Refactored `MnemosyneSessionState` to manage scoped recall/retain targets and deduplication.
- Updated hindsight tools to route recall/retain through scoped methods.
- Removed "ctx" suffix and cumulative Σ-token display from status lines.
- Replaced "N tools" text with tool count + extensionTool icon.
- Changed cost separator to ` . ` to visually distinguish it from dim stats.
- Added test asserting new format and absence of old labels.
can1357