Safer task flow with less manual coordination

Product stabilization adds runtime truth, trajectories, recovery, helper policy, an activity inspector, and fail-closed completion receipts without breaking existing workflows.

v1.6.1 · Reliability patch

Truthful UI state, safe external review, and lower helper overhead

This patch keeps the selected model and thinking level unchanged while tightening lifecycle truth, limiting helper expansion, and granting external-source reads only to the exact checkout for the current session.

Conversation truth

Drafts, retries, and pending completion stay in Activity. Chat preserves user-then-assistant order and clears processing only at the real terminal boundary.

External review

GitHub, GitLab, and Bitbucket references activate only minimal source checkout. The runtime grants reads to that exact tree while blocking caches, credentials, symlink escape, and every mutation path.

Solo-first

Direct subagent use is capped at one read-only helper with call, turn, and runtime budgets, no automatic retry/fork/worker expansion, and truthful failure state when evidence is missing.

Product stabilization

Six changes users will notice

Existing workflows continue to work. This update adds runtime evidence and fail-closed gates behind them, plus compact projections that users can inspect when needed.

Runtime truth

Model, provider, thinking, context window, and provenance come from the runtime or authenticated catalog. Missing facts stay unknown instead of being inferred from a model name.

Task trajectories

Tasks move through plan, execute, verify, review, handoff, and terminal using replayable state. Stale, corrupt, or identity-mismatched events fail closed.

Recovery and resume

Failures are classified before retry or repair. Environment, provider, permission, policy, and scope failures never grant source-mutation authority.

Role-bound helpers

Scouts, planners, reviewers, and Oracle receive bounded scope, tools, and budgets. The parent model stays pinned; automatic workers and multiple writers remain off.

Completion receipts

A receipt is approved only when the trace is completed, acceptance is satisfied, and the current-tree hard gate passes. Missing verifiers or pending work remain visible risks.

Product views

Preflight, live status, the activity inspector, efficiency, and orchestration policy use deterministic output; users do not need a model to remember or interpret runtime state.

Daily flow

The daily workflow does not break

Assign work as before

Use a natural-language request or /workflow task <request>. Manual preflight is optional.

The runtime records truth

Task Contracts, context, changed files, trajectories, and checkpoints are updated from hooks and tool results rather than model memory.

Verify the exact working tree

Only observed exact commands that exit successfully and match the current-tree digest count as passing evidence.

Bounded recovery or handoff

Source repair and transient retries have ceilings. Unsafe cases return to handoff or the operator instead of continuing by guesswork.

Completion fails closed

A completed trace is not enough. Work plan, acceptance, changed-file integrity, and current verifier evidence must pass before approval.

Inspect when needed

Five commands that expose the whole flow

These are inspection and diagnostic commands, not mandatory steps before every task.

Preflight, activity, status, metrics, and helper policy

/task-preflight --json Review current repository state
/piagent-inspector
/piagent-status
/usage efficiency
/piagent-orchestration

See How workflows change for exactly what each command reads, what it does not do, and when to use it.

Automatic vs opt-in

What runs automatically and what requires opt-in

Surface	Default	Behavior
Runtime snapshots and task evidence	Automatic	Records bounded facts, journal entries, verifier evidence, recovery, and receipts.
Activity inspector	Automatic four-row panel, optional menu	Shows task diffs, commands, safety, and context beside the footer without calling a model or changing workflow state.
Solver	shadow	Evaluates routes without changing the parent model or taking control of the workflow.
Phase tools	shadow	Computes projections; use PIAGENT_PHASE_TOOLS=on to apply phase-specific tool visibility.
Recovery	on	Retries or repairs within ceilings; use PIAGENT_AUTO_RECOVERY=off to restore ordinary handoff.
Helpers	off	Explicit PIAGENT_HELPERS_MODE=on still allows only one fresh read-only helper after two independent lanes and at least 30% projected net saving are proven.
Parent routing and automatic workers	off	No automatic parent-model switch, writer dispatch, or default multi-writer behavior.

Compatibility

No workflow rewrite or old-task migration required

Task Contract remains v2.

Runtime snapshots, solver decisions, trajectories, recovery, and helper state use additive versioned sidecars. Existing command aliases continue to work; old state without a new projection stays unknown or unavailable instead of being upgraded into success.

The host is not a sandbox

Policy reduces mistakes inside controlled tool surfaces but does not isolate code from the operator account. External and destructive actions still require confirmation.

Defaults remain conservative

Solver and phase tools remain in shadow, helpers do not auto-dispatch, and parent routing and automatic workers stay off until separate rollout evidence exists.

Try it

Test directly in a new Pi session

Read-only product walkthrough

/task-preflight --json Review this repository without modifying files
/piagent-inspector
/piagent-status
/usage efficiency

After updating the package or a local checkout, start a new Pi session so the extension reloads. See the Command catalog for every command and compatibility alias.