A single-model subagent is pinned to a `subagent:<id>` role whose
`retry.fallbackChains` entry shadows every configured role chain, so the
chain it inherits decides where the child retries. Inheritance resolved
the role by re-deriving it from the child's `modelPatterns` — but every
spawn path expands the role alias into `modelOverride` before calling
`runSubprocess` (`modelPatterns = normalizeModelPatterns(modelOverride ??
agent.model)`), so `@task` never reached the derivation and it returned
`undefined` every time. Every task subagent inherited `chains.default`.
With `modelRoles.task: anthropic/claude-sonnet-5`, `task` chained to
sonnet alone, and `default` chained to sonnet plus a second provider, a
transient stall on sonnet routed the child onto the default chain's
second model — one the operator had deliberately kept out of the `task`
chain — and a quota error there killed a 28-minute run.
#7694 fixed only the shape where an unexpanded alias reaches the
executor, which no production caller produces; its tests supplied a bare
`agent.model: ["@smol"]` with no `modelOverride`. The incident above
happened on v17.2.10, which contains that fix.
Route inheritance off the role identity the spawn path already computes
and passes as `modelRole`. Since that leaves the pattern-derived operand
unreachable, drop it and the parameter it was the only user of.
The vibe worker path had the same defect independently: `#resolveWorker`
expanded `@task`/`@smol` for the bundled `task`/`sonic` workers and kept
no role, so vibe children inherited `default` no matter what the
executor did. It now carries `modelRole` on `ResolvedVibeWorker` and
`VibeRecord` through both the spawn and rehydrate sites.
To stop the two halves drifting apart again — the mistake that caused
this bug — `resolveAgentModelSelection` returns the expanded `patterns`
and the pre-expansion `role` from one call, and both spawn paths take
both from it. `resolveAgentModelSource` is removed: its only use was
being fed to `resolveExplicitModelRole`, and keeping it invites the same
split derivation. `resolveAgentModelPatterns` stays for the UI callers
that legitimately want patterns alone.
Tests cover the producible shapes: the incident's chain layout (role
chain equal to the primary, default chain a superset), role identity
surviving expansion for every alias-routed bundled agent, and the
patterns/role pairing itself. #7694's two tests are re-anchored to a
shape a real caller produces.
The ExtensionAPI getAllTools() wired to session.getAllToolNames(),
returning bare tool-name strings. Upstream @earendil-works/pi-coding-agent
promises ToolInfo[] with sourceInfo, so extensions loaded through the
legacy-pi shim (e.g. gentle-pi) crashed on t.sourceInfo.source at every
session start.
Added SourceInfo/ToolInfo types plus SessionTools.getAllToolInfos(), which
returns { name, description, parameters, sourceInfo } and classifies each
tool as builtin/mcp/sdk/extension. Rewired every getAllTools action site
(interactive, acp, print/rpc, subagent executor) and the example extension.
Fixes#7732
A single-model subagent has no fallbacks of its own, so it inherits one
and is pinned to a `subagent:<id>` role. That pin is inserted first in
`retry.fallbackChains` so no other role can capture its routing — which
also means it shadows every configured role chain at runtime.
Inheritance was hardcoded to `chains.default`, so a subagent spawned
through a role alias (the bundled scout's `model: "@smol"`) retried on
the default role's chain instead of its own. With `smol` chained to
composer/grok/luna and `default` chained to gpt-5.6-sol, every scout
fell back onto sol.
Resolve the inherited chain from the role identity still present in the
raw pattern (`@smol` -> `smol`), falling back to `default` when that
role configures no chain. An explicitly empty role chain still means
"no fallbacks", mirroring `expandDefaultRetryFallbackChains`. Explicit
model selectors keep inheriting `default`: they carry no role identity,
and a role assigned the same model must not capture the child's routing.
Preserving aborted refs on dispose exposed a latent invariant break: the
executor's hard-abort path (finalizeSubagentLifecycle) set status `aborted`
and disposed the session without detaching it. With the ref now retained, it
kept a dangling pointer to the disposed session, and ensureLive returns any
non-null ref.session before its revivability check — so hub focus / transcript
chat could route into a dead session.
- finalizeSubagentLifecycle: detach the session before disposing on the
terminal hard-abort path, upholding the AgentRef invariant (session === null
when aborted).
- release(tombstone): detach before dispose too (capture the live session
first), same invariant.
- unregisterUnlessParked: preserve `aborted` refs only when already detached;
an aborted ref still holding a live session is a bug and is unregistered
rather than kept reachable.
- Regression test now asserts ensureLive rejects a tombstoned id as terminal.
Fixes#7250
- Extracted the shared attachIrcWakeTurnMonitor from the executor reviver closure.
- Installed it in the persisted cold-revive path, forwarding the top-level event bus.
- Covered that a resumed process's parked subagent emits wake lifecycle frames.
Fixes#7105
- Monitored autonomous IRC wake turns with the task executor lifecycle and progress channels.
- Preserved monitoring after idle-TTL parking and session revival.
- Covered RPC subscriptions for both idle and parked keep-alive agents.
Fixes#7105
The soft request budget resolved to `SOFT_REQUEST_BUDGET[agent.name] ??
configured`, so the bundled entries for scout and sonic replaced the
configured value outright. Lowering `task.softRequestBudget` to tighten
the guard therefore did nothing for exactly the two agents that spawn
most often: a scout kept its 100-request budget no matter how small the
user set the knob. Only 0 (disable) and raising the value for
non-bundled agents had any effect.
Treat both numbers as upper bounds and take the smaller one. The bundled
entries stay ceilings, so a runaway scout is still stopped at 100 by
default and existing behavior is unchanged for anyone who has not
lowered the setting; a configured 0 still disables the guard entirely.
Resolution moves into `resolveSoftRequestBudget`, which also normalizes
negative and fractional inputs, so the rule is testable without standing
up a subprocess run.
This composes with `task.maxEffort` on a separate axis: effort caps how
hard each request thinks, this caps how many requests a run may spend.
(cherry picked from commit f0db29f8f725f11390b64ca9342300c482ff5c5d)
- task.maxEffort only clamped the initial thinking level; a retry
fallback candidate could clamp back up to its model floor and run a
low-capped spawn at high.
- The ceiling now rides the session as thinkingLevelCeiling: clamped in
ModelControls (constructor, setThinkingLevel, auto classifier,
restore) and in applyRetryFallbackCandidate; fallback candidates whose
floor exceeds the ceiling are skipped.
- Effort value import moved to @oh-my-pi/pi-catalog/effort; changelog
attribution added.
- Review follow-up for PR #6794.
The prewalk arm/switch guard compared model identity only (modelsAreEqual /
provider+id), discarding the resolved thinkingLevel. A legal same-model target
at a cheaper effort (e.g. prewalk: "@task" resolving to the active model at a
lower level) was dropped as a no-op, so the session ran the expensive effort for
the whole run while still paying the plan/continue nudges — silently on the
session path, logger.debug only on the subagent path.
Compare (provider, id, effective thinking level) via a shared prewalkWouldBeNoop
helper. Effort-only deltas on the same model now switch; a genuine no-op emits a
user-visible notice on the session path and never arms on the subagent path.
Fixes#6659
- Threaded a resolvedModelIsFallback flag through AgentProgress and SingleResult.
- Set the flag from the executor retry-fallback handlers and settled results.
- Rendered the observer/no-session hub path as fallback -> provider/model.
- Added an observer-only fallback-badge regression test.
Fixes#6316
Shared the bundled task prewalk default between runtime execution and the Agent Control Center. Added dashboard regression coverage for task.prewalk.
Fixes#6306
Setting options.additionalDirectories to undefined wasn't enough:
createAgentSession also merges settings.get('workspace.additionalDirectories').
Now createSubagentSettings gets an override to clear the setting
when worktree is set, ensuring isolated runs can't edit outside the
worktree.
Co-authored-by: oh-my-pi <https://omp.sh>
Isolated tasks clone only cwd into the worktree. Forwarding the
parent's additional directories would let the subagent edit absolute
paths under the original extra roots, bypassing isolation. Now
additionalDirectories is undefined when worktree is set.
Co-authored-by: oh-my-pi <https://omp.sh>
Subagents (task tool) now inherit the parent session's
additionalDirectories via ToolSession → ExecutorOptions →
CreateAgentSessionOptions, so delegated agents see the same
<workspace-roots> block and can read/grep/glob added roots.
Co-authored-by: oh-my-pi <https://omp.sh>
MCP-backed tools never declared an explicit strict value, so OpenAI-family
serializers (post-#4336/#4340) had no false to preserve and models over-filled
mutually exclusive optional fields. Task/subagent proxies also rebuilt a raw
tools/call instead of executing through the source MCPTool, bypassing intent
stripping, placeholder pruning, local-URL resolution, reconnect, abort, and
result metadata; strict servers rejected proxied calls with
unrecognized_keys ["i"].
- MCPTool/DeferredMCPTool now declare `readonly strict = false as const`.
- createMCPProxyTools delegates to the current source tool, re-resolved by raw
MCP server/tool metadata so reconnect replacements are honored, and keeps the
Task 60s timeout by combining its abort signal with the caller's.
- Regression coverage: strict flags, proxy parity for i/placeholder shaping,
declared-i passthrough, and reconnect re-resolution.
Fixes#6208
Tracked active subagent session model changes in progress snapshots so prewalk handoffs replace the starting-model badge.
Added regression coverage for a prewalk handoff and documented the fix.
Fixes#6083
The barrier in driveSessionToYield was unreachable for terminal yields:
the yield tool's shouldTerminate fired requestAbort("terminate"), so
abortSignal was always aborted before the barrier's loop condition ran,
and a run with pending owner jobs completed immediately with whatever
the pre-async yield said (Codex review on #6119).
- Split "stop the free-running turn after yield" from "terminate the
run": a terminal yield with pending owner async work now parks the
run with a recoverable session abort (budget-stop precedent) via
requestYieldTurnStop; only a quiescent yield terminates.
- An async-result follow-up injected after a recorded yield un-latches
it (transcript-ordered, in the run monitor) and re-runs the reminder
ladder, so the run only completes on a yield that postdates every
delivered result — including results injected during the notice turn.
- A run that never refreshes a superseded yield fails (exit 1) with an
explicit reason; the stale payload ships only as failed-run salvage
through the existing failed-after-yield finalize path.
- Rewrote subagent-async-pending.md: the "your current yield stands"
option contradicted the enforced contract.
- Regression tests: parked yield -> injected result -> fresh yield wins;
refusal -> stale payload fails; no-async fast path unchanged.
Narrow the invalid-yield guard to !abortSent so array-typed incremental
yield sections no longer suppress the infinite-submit-loop abort; add
regression coverage for incremental yield followed by repeated malformed
terminal yields.
can1357