@bastani/atomic 0.9.18 → 0.9.19-alpha.2

package/dist/builtin/web-access/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/web-access",
-  "version": "0.9.18",
+  "version": "0.9.19-alpha.2",
   "private": true,
   "description": "Atomic extension for web search, URL fetching, GitHub repo cloning, PDF/video extraction. Fork of: https://github.com/nicobailon/pi-web-access",
   "contributors": [
@@ -10,7 +10,7 @@
   "license": "MIT",
   "type": "module",
   "engines": {
-    "bun": ">=1.4.0"
+    "bun": ">=1.4.2"
   },
   "main": "./index.ts",
   "types": "./index.ts",

package/dist/builtin/workflows/CHANGELOG.md

@@ -6,6 +6,54 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 
 ## [Unreleased]
 
+## [0.9.19-alpha.2] - 2026-09-08
+
+### Added
+
+- Connected workflow activity to the host extension observer stream, independently of lifecycle-notification settings and attribution filters. Root activity is projected from workflow snapshots plus run-qualified runtime execution ownership: nested runs fold into full root replacements; independent execution is distinguished from human waits; runnable handoffs, retries, stop draining, pauses, and acknowledged failures are accounted for without treating historical running stages as execution. Tool-only execution, parallel human-input waits, pause, cancellation drain, and unresolved failures publish root activity replacements; late attachment receives current state, and durable hydration announces recovering before ready. Typed lifecycle hooks cover run, stage, tool, prompt, and control transitions with canonical nested identities; successful-stage completion hooks exclude failed/skipped outcomes, explicit execution replay is tagged, and restored history creates no synthetic completions. Heartbeat hooks follow the existing configured cadence without adding timers or graph nodes. Because the workflows extension publishes root activity to the host, Atomic's built-in Herdr reporter reflects workflow execution and human-input waits in the owning pane; the end-to-end path is covered by an integration test against a fake Herdr CLI ([#2891](https://github.com/bastani-inc/atomic/issues/2891)).
+- Attached stage chats now suggest stage-local `/skill:` commands with source metadata, qualified selectors, and refreshed catalogs after resource reload. Idle, steering, and follow-up submissions reuse session expansion once, retain stage admission and HIL ownership, and show skill diagnostics locally. Explicit editable post-mortem chats can invoke skills without reopening workflow execution.
+- `StageSendUserMessageOptions.expandPromptTemplates` opts native stage-session delivery into existing command and skill/template expansion; ordinary programmatic messages remain literal by default.
+- Attached stage chats render a bound owner task store through the shared chat host, keeping task activity separate from pending launch-tool replay and retaining the compact task footer after the launch row scrolls away. This requires the stage producer to bind its owner store ([#2884](https://github.com/bastani-inc/atomic/pull/2884)).
+
+### Changed
+
+- Goal and Ralph orchestration, Ralph research, and Open Claude Design now use GPT-6 Astra at medium. Prompt refinement stays at high, and OpenRouter Grok fallbacks explicitly use xhigh.
+- `/tasks` in an attached stage chat opens the shared owner-bound inspector without entering model context, including during interrupt settlement. Task focus exits before the ordinary stage Escape action, and mounted human-input prompts retain input priority ([#2884](https://github.com/bastani-inc/atomic/pull/2884)).
+- Stage chat now shows compact background-agent and shell counts below the composer instead of persistent task rows. Shared `/tasks` views retain background results, and completion notifications use shaded cards with readable outcomes and previews in the owning stage chat.
+- Workflow guidance now uses project CI/test timing history for critical-path scheduling and launch estimates, avoids redundant full-suite runs while preserving required gates, and compares estimated with actual elapsed time. Agent-authored user questions use `ask_user_question` or an equivalent tool when available; sessions without one continue autonomously using best judgment.
+- Workflow guidance treats requests to work "quickly" as task-scoped inline execution. Duration estimates appear before launch without confidence labels. Launches inherit configured limits without a routine budget-choice question; explicit user limits and approval before raising an exhausted budget remain enforced.
+
+### Fixed
+
+- Paused stage skill submissions retain Enter versus Ctrl+F delivery intent. Custom stage hosts without admission-aware user-message delivery now report invocation as unavailable rather than using an unguarded fallback.
+- Skill autocomplete reuses the attached stage session without writing durable checkpoints per keystroke. Concurrent lazy discovery requests share one attachment.
+- Same-name prompt/tool nodes no longer cause valid duplicate agent matches to be refused as non-agents. Live `ask` retains ambiguity diagnostics and name-based `send` retains sticky agent delivery ([#2895](https://github.com/bastani-inc/atomic/pull/2895)).
+- Keep a bound task footer visible beside mounted stage questions within the existing viewport budget, and apply the host's expansion setting consistently to task rows.
+- Restored `/tasks` to stage slash suggestions and removed filled backgrounds from skill suggestions, including selected rows. Task status now follows MCP, retaining active counts and `/tasks` without collision with the graph-return shortcut on narrow terminals.
+- Stage pause keeps owned background agents and shells running. Closing the stage generation cancels that owner's remaining work without stopping sibling stages.
+- Preserve an ordinary stage's complete output artifact when a follow-up clarification arrives before completion. Completed answers from the same prompt generation remain ordered supplements, including executor continuations and close-time delivery, without copying earlier history or tool progress. Accepted answers survive context compaction and model fallback during a continuation; failed attempts cannot discard an earlier successful report. Tree navigation cannot import historical answers, and generation close stops capture of later retained-session chat. Completed artifact receipts stay stable under repeated finalization; structured output keeps its existing capture contract.
+- Discard provisional artifact answers from each failed same-model retry, preserving earlier accepted reports and only publishing the successful retry's answers.
+- Prevent workflow activity from briefly reporting idle between settled nodes and live author continuation. Historical, paused, blocked, and stopping runs do not gain continuation ownership ([#2891](https://github.com/bastani-inc/atomic/issues/2891)).
+- Preserved workflow lifecycle `kill` events for already-aborted caller signals and report task-result checkpoint interrupts as `interrupt`, without changing their graceful paused outcome ([#2912](https://github.com/bastani-inc/atomic/pull/2912)).
+- Durable tool callbacks now wait for agent tasks they admit to reach terminal results before checkpointing. Yielded observations are replaced with terminal observations without changing the cancellation-before-persistence or commit-wins fences ([#2884](https://github.com/bastani-inc/atomic/pull/2884)).
+- Workflow graph canvas and node interiors now use the terminal's background like main and stage chat, instead of painting a fixed dark palette. Themed chrome and focused title tabs remain unchanged.
+- Graph chrome and `/workflow connect` picker rows retain solid background fill across truncated names, filters, and hints. Truncated stage and child-workflow labels keep their focused tab's background, text color, and weight through the ellipsis. Skill, command, and file autocomplete retain terminal-default backgrounds, including selected rows.
+- Classified the workflow graph and run picker as navigation so `/workflow connect` no longer creates a false Herdr approval wait. Actual workflow input waits remain reported.
+- Workflow-stage `/tasks` now retains the shared compact picker and the stage's model, reasoning, cwd/branch, and MCP footer. Detail pages stay fullscreen. Task navigation no longer triggers the main-chat input notice, and pending-prompt notices clear when the graph hides, closes, or changes host.
+- Resume ended recoverable blocks through the existing continuation path rather than returning an unchanged snapshot as success. Non-resumable targets and unchanged blocked snapshots report no progress.
+- Settled failed or blocked runs retain attention without reporting an active user-decision wait to Herdr. Pending prompts and exhausted budgets remain blocked.
+- Stopped the workflow graph projection from retaining and deep-freezing a completed stage's structured result, and gave input defaulting a resolver-owned copy, so a stage's structured output can still be handed to a typed child workflow ([#2936](https://github.com/bastani-inc/atomic/issues/2936)).
+
+## [0.9.19-alpha.1] - 2026-09-06
+
+### Fixed
+
+- Interrupting a workflow tool call during initialization now cancels its startup owner instead of leaving an empty, uncontrollable running root. Resume preparation respects cancellation, and already-aborted requests cannot begin actions. Accepted request timeouts and acknowledged background execution retain their existing behavior.
+- Whole-run pause/interrupt during initialization or between-node awaits now retain the same live executor for explicit resume instead of becoming quit. Later tracked work, cached replay, and completion wait for resume; untracked JavaScript or I/O may still finish. Executor-only quit retires the owner, and zero-progress quits remain nonresumable. Cross-process resume still requires durable progress.
+- Executor-only whole-run pause now holds already-live nested child owners as well, preventing their next tracked step or completion until resume. Descendants remain held while the root's durable resume is acknowledged; child-scoped controls preserve sibling isolation.
+- Preserved the original selected tool failure when cancellation arrives later, restored closed-admission errors for retained tool calls, and drained cancelled child boundaries before the parent returns. Durable resume replaces abandoned detached executors without replacing surviving live paused owners.
+- Fixed embedded PostgreSQL discovery in Bun-compiled installations and ordinary native npm leaves, including nested installs, while retaining explicit database/runtime overrides and legacy package fallback.
+
 ## [0.9.18] - 2026-09-05
 
 Cumulative release of the `0.9.18-alpha.3` through `0.9.18-alpha.7` prereleases. Per-change details remain in the unchanged prerelease sections below.

package/dist/builtin/workflows/README.md

@@ -689,7 +689,7 @@ Raw stage-chat prompt answer replay is live-memory only. `StageSnapshot.promptAn
 ```json
 {
   "name": "workflow",
-  "description": "Run named builtin, project, user, or package workflows; custom definitions may import reusable project/package workflows or builtin definitions from @bastani/atomic/workflows/builtin and nest them with ctx.workflow(...), including deeper composition within the configured maxDepth; when workflow execution fits but another shape would better achieve the task, author a custom TypeScript workflow({...}) inline with normal coding tools, reload it, and run it; after successfully creating and reloading a newly authored custom workflow, report the folder containing its generated code as 'Custom workflow created. You can inspect its code at: <workflow-folder-path>'; do this only for newly created custom workflows, never builtin or pre-existing workflows; discover with list/get/inputs/models, list session runs with status (no runId; statusFilter narrows the list), inspect status/stages/stage details, answer pending prompts only while the root workflow is nonterminal, pause/resume/interrupt/quit runs, and reload workflow resources. When steering or communication is useful, use Intercom. Before steering a stage, join its invocation group. Use the Intercom `groups` action to discover it. Workflow invocation groups are named `workflow:<rootRunId>`. Then address the stage as `workflow:<rootRunId>/<segment>[/<segment>...]`; each segment may be a stage name, run id, or glob (`*` matches one segment and `**` any depth). Use `intercom list` inside the invocation group to see live, pending, and possible future targets. Live delivery is immediate; a known stage that has not started is queued and receives the message before its first model turn. Name and pattern sends remain sticky for every future matching stage until the root run terminates. When shared scope or acceptance criteria change, broadcast one authoritative update to `workflow:<rootRunId>/**` (or a narrower path pattern) rather than enumerating stages. A syntactically valid target outside the known set queues with a `notInKnownSet` warning and settles undeliverable at terminal only if never delivered. Use `ask` only for a reply-capable live session. Treat a blocked run as continuable by default unless the user requests inline/no-workflow execution: then safely hold/stop it, reconcile completed work and in-flight side effects, and continue inline without duplication. Otherwise resume resumable blocks, answer pending prompts, steer past the obstacle, or start a follow-up workflow past a terminal block; resolve material ambiguity from user intent and repository evidence without bypassing safety or authorization. For action 'run' and 'resume', budget accepts per-field duration, token, cost, and warning overrides; fields resolve over the workflow declaration and config, and 0 disables a field. Pass budget only when the user asked for a limit; otherwise omit it entirely and inherit the declaration and config rather than inventing a cap. For primitive prompt answers, use booleans or the documented confirm labels, exact case-insensitive select labels or 1-based indexes, and text strings for input/editor; an invalid answer remains pending and returns guidance instead of choosing a default. For large stage handoffs, write context to files/artifacts, pass paths via reads, and prompt downstream agents to 'Read the file at <path>...' instead of injecting large previous text. Wrap critical parts of run inputs and steering messages in <keepContext>...</keepContext> so compaction preserves them verbatim in the stages that inherit them; tag role constraints, prohibitions, must-hold criteria, and identifiers, not background or bulk reference material. For transcripts, prefer status/stages/stage to get sessionFile/transcriptPath, quote the exact path without rewriting separators (Windows backslashes are valid), then search it with rg/grep and read small ranges; transcript is path-only by default when sessionFile/transcriptPath exists, explicit tail/limit returns bounded previews, and missing transcript paths fall back to a small preview. Use action 'models' to inspect models in the configured catalog; the result is a configured-auth snapshot showing what's present in the registry with configured authentication, not proof of credentials, entitlements, OAuth freshness, or live provider access. When authoring a workflow that should dynamically select a model, first call workflow({ action: 'models' }) to inspect the configured catalog, then select from the returned provider/id entries considering the isCurrent marker and available thinking levels.",
+  "description": "Run named builtin, project, user, or package workflows; custom definitions may import reusable project/package workflows or builtin definitions from @bastani/atomic/workflows/builtin and nest them with ctx.workflow(...), including deeper composition within the configured maxDepth; Requests to do the task 'quickly' or 'inline' mean no workflow creation or launch, including hidden/nested replacements. when workflow execution fits but another shape would better achieve the task, author a custom TypeScript workflow({...}) inline with normal coding tools, reload it, and run it; after successfully creating and reloading a newly authored custom workflow, report the folder containing its generated code as 'Custom workflow created. You can inspect its code at: <workflow-folder-path>'; do this only for newly created custom workflows, never builtin or pre-existing workflows; discover with list/get/inputs/models, list session runs with status (no runId; statusFilter narrows the list), inspect status/stages/stage details, answer pending prompts only while the root workflow is nonterminal, pause/resume/interrupt/quit runs, and reload workflow resources. When steering or communication is useful, use Intercom. Before steering a stage, join its invocation group. Use the Intercom `groups` action to discover it. Workflow invocation groups are named `workflow:<rootRunId>`. Then address the stage as `workflow:<rootRunId>/<segment>[/<segment>...]`; each segment may be a stage name, run id, or glob (`*` matches one segment and `**` any depth). Use `intercom list` inside the invocation group to see live, pending, and possible future targets. Live delivery is immediate; a known stage that has not started is queued and receives the message before its first model turn. Name and pattern sends remain sticky for every future matching stage until the root run terminates. When shared scope or acceptance criteria change, broadcast one authoritative update to `workflow:<rootRunId>/**` (or a narrower path pattern) rather than enumerating stages. A syntactically valid target outside the known set queues with a `notInKnownSet` warning and settles undeliverable at terminal only if never delivered. Use `ask` only for a reply-capable live session. Treat a blocked run as continuable by default unless the user requests inline/no-workflow execution: then safely hold/stop it, reconcile completed work and in-flight side effects, and continue inline without duplication. Otherwise resume resumable blocks, answer pending prompts, steer past the obstacle, or start a follow-up workflow past a terminal block; resolve material ambiguity from user intent and repository evidence without bypassing safety or authorization. For action 'run' and 'resume', budget accepts per-field duration, token, cost, and warning overrides; fields resolve over the workflow declaration and config, and 0 disables a field. Before launching, share an estimated duration without confidence labels. Proceed with inherited budget limits without asking the user to choose a budget before each launch. Pass budget only for a user-specified limit; otherwise omit it and inherit the declaration and config. An estimate is not a budget. For primitive prompt answers, use booleans or the documented confirm labels, exact case-insensitive select labels or 1-based indexes, and text strings for input/editor; an invalid answer remains pending and returns guidance instead of choosing a default. For large stage handoffs, write context to files/artifacts, pass paths via reads, and prompt downstream agents to 'Read the file at <path>...' instead of injecting large previous text. Wrap critical parts of run inputs and steering messages in <keepContext>...</keepContext> so compaction preserves them verbatim in the stages that inherit them; tag role constraints, prohibitions, must-hold criteria, and identifiers, not background or bulk reference material. For transcripts, prefer status/stages/stage to get sessionFile/transcriptPath, quote the exact path without rewriting separators (Windows backslashes are valid), then search it with rg/grep and read small ranges; transcript is path-only by default when sessionFile/transcriptPath exists, explicit tail/limit returns bounded previews, and missing transcript paths fall back to a small preview. Use action 'models' to inspect models in the configured catalog; the result is a configured-auth snapshot showing what's present in the registry with configured authentication, not proof of credentials, entitlements, OAuth freshness, or live provider access. When authoring a workflow that should dynamically select a model, first call workflow({ action: 'models' }) to inspect the configured catalog, then select from the returned provider/id entries considering the isCurrent marker and available thinking levels.",
   "parameters": {
     "workflow": "string (optional) — workflow ID or normalized name",
     "inputs": "object (optional) — key/value map of workflow inputs",

package/dist/builtin/workflows/builtin/adversarial-verification.js

@@ -1,12 +1,10 @@
 import {
   adversarial_verification_default
-} from "./chunk-reqr3sr3.js";
-import"./chunk-7at6dnkr.js";
-import"./chunk-7430zyas.js";
-import"./chunk-wpckd35c.js";
-import"./chunk-cg9tmks0.js";
-import"./chunk-bfkmzv9h.js";
-import"./chunk-zbekjqew.js";
+} from "./chunk-1kx4aggw.js";
+import"./chunk-p8jjwfg6.js";
+import"./chunk-n1910xc4.js";
+import"./chunk-xtxcq5tj.js";
+import"./chunk-a2kdsqn8.js";
 export {
   adversarial_verification_default as default
 };

package/dist/builtin/workflows/builtin/{chunk-reqr3sr3.js → chunk-1kx4aggw.js}

@@ -1,26 +1,26 @@
 import {
-  fold_usage
-} from "./chunk-7at6dnkr.js";
-import {
-  build_scoring_prompt,
-  scoring_prompt_reads,
-  warm_first_fan_out
-} from "./chunk-7430zyas.js";
+  workflow
+} from "./chunk-xtxcq5tj.js";
 import {
-  stableArtifactRoot
-} from "./chunk-wpckd35c.js";
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
 import {
-  VERIFICATION_SCALE,
-  decide_verification,
+  parse_rubric,
   normalize_criteria,
-  parse_rubric
-} from "./chunk-cg9tmks0.js";
+  VERIFICATION_SCALE,
+  decide_verification
+} from "./chunk-n1910xc4.js";
 import {
-  workflow
-} from "./chunk-bfkmzv9h.js";
+  scoring_prompt_reads,
+  build_scoring_prompt,
+  warm_first_fan_out
+} from "./chunk-p8jjwfg6.js";
 import {
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  fold_usage
+} from "./chunk-tgt0s5e5.js";
+import {
+  stableArtifactRoot
+} from "./chunk-rc5ec5xx.js";
 
 // dist/builtin/workflows/builtin/adversarial-verification.ts
 import { Type as Type2 } from "typebox";
@@ -258,10 +258,10 @@ async function runAdversarialVerification(ctx) {
         });
       }
     }
-    const runWave = async (pending2, reaskWave) => {
-      if (pending2.length === 0)
+    const runWave = async (pending, reaskWave) => {
+      if (pending.length === 0)
         return { valid: [], invalid: [], results: [] };
-      const reports = await warm_first_fan_out(ctx, pending2.map((cell) => scoreStep(cell, reaskWave, scoringHead, criteriaPath)), () => scoringHead, { concurrency: Math.min(pending2.length, 4), failFast: false });
+      const reports = await warm_first_fan_out(ctx, pending.map((cell) => scoreStep(cell, reaskWave, scoringHead, criteriaPath)), () => scoringHead, { concurrency: Math.min(pending.length, 4), failFast: false });
       const byName = new Map;
       for (const report of reports) {
         const name = report.name ?? report.stageName;
@@ -270,7 +270,7 @@ async function runAdversarialVerification(ctx) {
       }
       const valid = [];
       const invalid = [];
-      for (const cell of pending2) {
+      for (const cell of pending) {
         const report = byName.get(stageName(cell, reaskWave));
         const artifactPath = reaskWave === 0 ? cell.artifactPath : cell.artifactPath.replace(/\.json$/, `-reask-${reaskWave}.json`);
         if (isCriterionScore(report?.structured, cell.criterion.id)) {
@@ -398,7 +398,7 @@ var adversarial_verification_default = workflow({
     review_report_path: Type2.String({ description: "Path to the final consolidated findings or quorum report." }),
     remaining_work: Type2.Array(Type2.String(), { description: "Unresolved findings or quorum evidence when not approved." })
   },
-  run: async (ctx) => await runAdversarialVerification(withSteeringPropagationContext(ctx))
+  run: async (ctx) => await runAdversarialVerification(withSteeringPropagationContext2(ctx))
 });
 
 export { adversarial_verification_default };

package/dist/builtin/workflows/builtin/{chunk-42s7sw0b.js → chunk-62v3gg9f.js}

@@ -1,9 +1,9 @@
 import {
   workflow
-} from "./chunk-bfkmzv9h.js";
+} from "./chunk-xtxcq5tj.js";
 import {
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
 
 // dist/builtin/workflows/builtin/open-claude-design.ts
 import { Type as Type2 } from "typebox";
@@ -806,10 +806,10 @@ async function runOpenClaudeDesignWorkflow(ctx) {
     }
   };
   const designModelConfig = {
-    model: "openai-codex/gpt-6-astra:high",
+    model: "openai-codex/gpt-6-astra:medium",
     fallbackModels: [
-      "github-copilot/gpt-6-astra:high",
-      "openai/gpt-6-astra:high",
+      "github-copilot/gpt-6-astra:medium",
+      "openai/gpt-6-astra:medium",
       "anthropic/claude-fable-5-1:high",
       "github-copilot/claude-fable-5-1:high",
       "anthropic/claude-opus-5:high",
@@ -832,7 +832,7 @@ async function runOpenClaudeDesignWorkflow(ctx) {
       "zai-coding-cn/glm-5.3-flash:high",
       "baseten/zai-org/GLM-5.3:high",
       "baseten/zai-org/GLM-5.3-Flash:high",
-      "openrouter/openai/gpt-6-astra:high",
+      "openrouter/openai/gpt-6-astra:medium",
       "openrouter/anthropic/claude-fable-5-1:high",
       "openrouter/anthropic/claude-opus-5:high",
       "openrouter/anthropic/claude-fable-5:high",
@@ -840,7 +840,7 @@ async function runOpenClaudeDesignWorkflow(ctx) {
       "openrouter/moonshotai/kimi-k3:max",
       "openrouter/openai/gpt-5.6-sol:xhigh",
       "openrouter/sakana/fugu-ultra:high",
-      "openrouter/x-ai/grok-4.6",
+      "openrouter/x-ai/grok-4.6:xhigh",
       "openrouter/z-ai/glm-5.3:high",
       "openrouter/z-ai/glm-5.3-flash:high"
     ]
@@ -1066,7 +1066,7 @@ var open_claude_design_default = workflow({
     spec_file_url: Type2.Optional(Type2.String({ description: "file:// URL for the generated spec.html file." })),
     playwright_cli_status: Type2.Optional(Type2.String({ description: "Outcome of the initial deterministic step that ensures the playwright-cli skill's `playwright-cli` command is installed." }))
   },
-  run: async (ctx) => await runOpenClaudeDesignWorkflow(withSteeringPropagationContext(ctx))
+  run: async (ctx) => await runOpenClaudeDesignWorkflow(withSteeringPropagationContext2(ctx))
 });
 
 export { open_claude_design_default };

package/dist/builtin/workflows/builtin/{chunk-wgccch41.js → chunk-6w6cvk6b.js}

@@ -1,16 +1,16 @@
 import {
-  stableArtifactRoot
-} from "./chunk-wpckd35c.js";
+  workflow
+} from "./chunk-xtxcq5tj.js";
 import {
-  classify_trend,
-  score_progress
-} from "./chunk-6fqs7c01.js";
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
 import {
-  workflow
-} from "./chunk-bfkmzv9h.js";
+  score_progress,
+  classify_trend
+} from "./chunk-z0d03ktm.js";
 import {
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  stableArtifactRoot
+} from "./chunk-rc5ec5xx.js";
 
 // dist/builtin/workflows/builtin/loop-until-done.ts
 import { Type as Type2 } from "typebox";
@@ -246,7 +246,7 @@ async function runLoopUntilDone(ctx) {
     }
     await writeLedger(ledgerPath, ctx.inputs.prompt, ctx.inputs.max_iterations, decision.done ? "complete" : "active", entries);
     if (decision.done) {
-      const report2 = progressReport(entries);
+      const report = progressReport(entries);
       const resultPath = join(artifactDir, "result.md");
       const final = await ctx.task("completion-summary", {
         prompt: renderCompletionPrompt({ task: ctx.inputs.prompt, ledgerPath, iterationPath }),
@@ -255,12 +255,12 @@ async function runLoopUntilDone(ctx) {
         output: resultPath,
         outputMode: "file-only"
       });
-      const result2 = `${final.text ? `${final.text.trimEnd()}
+      const result = `${final.text ? `${final.text.trimEnd()}
 
-` : ""}${formatProgressReport(report2)}
+` : ""}${formatProgressReport(report)}
 `;
       return {
-        result: result2,
+        result,
         status: "complete",
         iterations_completed: iteration,
         ledger_path: ledgerPath,
@@ -269,8 +269,8 @@ async function runLoopUntilDone(ctx) {
         result_path: resultPath,
         remaining_work: "",
         artifact_dir: artifactDir,
-        progress_curve: report2.curve,
-        final_trend: report2.trend,
+        progress_curve: report.curve,
+        final_trend: report.trend,
         progress_disclaimer: PROGRESS_DISCLAIMER
       };
     }
@@ -339,7 +339,7 @@ var loop_until_done_default = workflow({
     final_trend: Type2.Union([Type2.Literal("rising"), Type2.Literal("flat"), Type2.Literal("regressing")], { description: "Final advisory trend classification; never a stop decision." }),
     progress_disclaimer: Type2.String({ description: "Calibration disclaimer for the advisory progress signal." })
   },
-  run: async (ctx) => await runLoopUntilDone(withSteeringPropagationContext(ctx))
+  run: async (ctx) => await runLoopUntilDone(withSteeringPropagationContext2(ctx))
 });
 
 export { loop_until_done_default };

package/dist/builtin/workflows/builtin/{chunk-zbekjqew.js → chunk-a2kdsqn8.js}

@@ -22,7 +22,7 @@ var PENDING_STAGE_DELIVERY_GUIDANCE = PENDING_STAGE_DELIVERY_GUIDANCE_LINES.join
 var STEERING_PROPAGATION_CONTRACT = [
   "Steering propagation contract:",
   "- Mid-run user messages (steering, follow-ups, resume text) are authoritative and may amend this run's objective or acceptance criteria. Adopt an amendment as required behavior from the moment you receive it.",
-  `- Explicit task-scoped requests to work "inline", "do this directly", or "don't use a workflow" override workflow-first defaults even for complex work. Do not create or launch a hidden/nested replacement workflow or pressure the user to reapprove the choice. Quoted examples and questions about inline code are not execution-mode instructions.`,
+  `- Explicit task-scoped requests to work "quickly", "inline", "do this directly", or "don't use a workflow" override workflow-first defaults even for complex work. Treat "quickly" as an inline execution choice, not a request for a faster workflow. Do not create or launch a hidden/nested replacement workflow or pressure the user to reapprove the choice. Quoted examples and questions about inline code are not execution-mode instructions; neither are descriptions of software that should run quickly.`,
   "- On a switch from an active workflow to inline, notify the controlling session to safely hold/stop the affected workflow, reconcile completed work and in-flight side effects, then continue inline without duplicate execution. Do not claim completed work was undone. Preserve appropriate tests, review, evidence, safety and authorization constraints; the preference applies only to the specified task.",
   "- An amendment that stays in your session is lost. Restate every objective-relevant steering message in your final report or handoff artifact, verbatim when short, under an explicit `Contract amendments received` heading.",
   "- Keep user-authored amendments visibly separate from your own observations, so later stages can tell a required clause from an agent proposal.",
@@ -64,7 +64,7 @@ var WORKER_PREFLIGHT_CONTRACT = [
 var E2E_VERIFICATION_GUIDANCE = [
   "Verify correctness end-to-end whenever practical for user-visible behavior; an executable scenario is stronger proof than code inspection, unit tests, or stage summaries alone.",
   'For web or frontend flows, including frontend changes whose correctness depends on backend/API behavior, use the playwright-cli skill or delegate with `skill: "playwright-cli"`; capture snapshot, screenshot, DOM, or network evidence suited to the objective.',
-  'For TUI/terminal flows, use the tmux skill or delegate with `skill: "tmux"`; select tmux on supported hosts, native Windows psmux when available, or herdr\'s pane API when appropriate. Check installed help: psmux uses tmux-style commands, while herdr has its own CLI, not interchangeable flags. Exercise the actual interactive flow and retain pane/text or terminal recording evidence.',
+  "For TUI/terminal automation/testing, prefer the herdr skill on macOS, Linux and Windows. Install Herdr if missing when network access and permissions permit; fall back to the tmux skill or native Windows psmux if installation or use is not possible. Preserve the pinned herdr skill's explicit-request and HERDR_ENV=1 requirements; never control a focused session from outside Herdr. Check installed help: psmux uses tmux-style commands, while herdr has its own CLI, not interchangeable flags. Exercise the actual interactive flow and retain pane/text or terminal recording evidence.",
   "For desktop and accessible simulator/emulator windows, use suitable PyAutoGUI or native platform tooling for a real visible app scenario and screenshots/video. Consult https://github.com/openai/openai-cua-sample-app#first-run and its Python README for setup and interruption guidance; the sample is optional and needs model/API access. Use a dedicated desktop/session, controlled fixtures and required screen-recording/accessibility permissions, inspect screenshots for other windows, and release held keys/buttons on interruption. Do not automate an arbitrary physical iPhone without an accessible simulator/emulator/mirroring or native automation setup, or label browser recordings as terminal/iOS proof.",
   "Choose tools from observed environment capabilities. Prefer installed/cached tools; install missing tools, runtimes or plugins when online access and permissions permit. Known offline/restricted installation is sufficient evidence not to attempt prohibited downloads; otherwise check capabilities or make one bounded setup attempt. Do not loop on blocked downloads, require production credentials, or take unsafe actions.",
   "When execution or installation is unavailable, continue available authoritative repository checks and retain useful local evidence. Report the constraint or attempted command with observed output, what could not run, and the narrower validation performed. For non-UI tasks, use relevant executable checks rather than manufacturing a UI scenario. An unavailable optional mechanism is not itself failed implementation or permission to relax required project checks."
@@ -204,7 +204,7 @@ function withStepContract(step) {
   const withPrompt = step.prompt === undefined ? step : { ...step, prompt: withSteeringPropagation(step.prompt) };
   return withPrompt.task === undefined ? withPrompt : { ...withPrompt, task: withSteeringPropagation(withPrompt.task) };
 }
-function withSteeringPropagationContext(ctx) {
+function withSteeringPropagationContext2(ctx) {
   const wrapped = Object.create(ctx);
   wrapped.task = (name, options) => ctx.task(name, withPromptContract(options));
   wrapped.chain = (steps, options) => ctx.chain(steps.map(withStepContract), options);
@@ -212,4 +212,4 @@ function withSteeringPropagationContext(ctx) {
   return wrapped;
 }
 
-export { keepContext, REVIEWER_CALIBRATION_RULES, WORKER_PREFLIGHT_CONTRACT, E2E_VERIFICATION_GUIDANCE, CODE_QUALITY_VERIFICATION_GUIDANCE, MEDIA_PUBLICATION_GUIDANCE, REPO_INTENT_MINING_GUIDANCE, renderE2eQaVideoReviewGuidance, LITERAL_OBJECTIVE_CONTRACT, REVIEWER_SPEC_VS_OBJECTIVE_GUARD, REVIEWER_OVERIMPLEMENTATION_GUARD, ACCEPTANCE_MATRIX_CONTRACT, CONTRACT_FIDELITY_AUDIT, REVIEWER_INTERCOM_COORDINATION_PROTOCOL, REVIEWER_INDEPENDENT_VERIFICATION_CONTRACT, REGRESSION_EVIDENCE_CONTRACT, FINDINGS_CONSOLIDATION_CONTRACT, SCOPE_DISCIPLINE_CONTRACT, EVIDENCE_CLOSURE_POLICY, WORKTREE_DISCIPLINE_CONTRACT, REVIEW_CODE_DELTA_CONTRACT, withSteeringPropagationContext };
+export { keepContext, REVIEWER_CALIBRATION_RULES, WORKER_PREFLIGHT_CONTRACT, E2E_VERIFICATION_GUIDANCE, CODE_QUALITY_VERIFICATION_GUIDANCE, MEDIA_PUBLICATION_GUIDANCE, REPO_INTENT_MINING_GUIDANCE, renderE2eQaVideoReviewGuidance, LITERAL_OBJECTIVE_CONTRACT, REVIEWER_SPEC_VS_OBJECTIVE_GUARD, REVIEWER_OVERIMPLEMENTATION_GUARD, ACCEPTANCE_MATRIX_CONTRACT, CONTRACT_FIDELITY_AUDIT, REVIEWER_INTERCOM_COORDINATION_PROTOCOL, REVIEWER_INDEPENDENT_VERIFICATION_CONTRACT, REGRESSION_EVIDENCE_CONTRACT, FINDINGS_CONSOLIDATION_CONTRACT, SCOPE_DISCIPLINE_CONTRACT, EVIDENCE_CLOSURE_POLICY, WORKTREE_DISCIPLINE_CONTRACT, REVIEW_CODE_DELTA_CONTRACT, withSteeringPropagationContext2 };

package/dist/builtin/workflows/builtin/{chunk-ze5x3d1r.js → chunk-fg9esgkj.js}

@@ -1,12 +1,12 @@
-import {
-  stableArtifactRoot
-} from "./chunk-wpckd35c.js";
 import {
   workflow
-} from "./chunk-bfkmzv9h.js";
+} from "./chunk-xtxcq5tj.js";
 import {
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
+import {
+  stableArtifactRoot
+} from "./chunk-rc5ec5xx.js";
 
 // dist/builtin/workflows/builtin/fan-out-and-synthesize.ts
 import { Type as Type2 } from "typebox";
@@ -193,7 +193,7 @@ var fan_out_and_synthesize_default = workflow({
     artifact_dir: Type2.String({ description: "Per-run artifact directory." }),
     manifest_path: Type2.String({ description: "Barrier manifest linking partitions to branch artifacts." })
   },
-  run: async (ctx) => await runFanOutAndSynthesize(withSteeringPropagationContext(ctx))
+  run: async (ctx) => await runFanOutAndSynthesize(withSteeringPropagationContext2(ctx))
 });
 
 export { fan_out_and_synthesize_default };

package/dist/builtin/workflows/builtin/{chunk-n8h4yyew.js → chunk-g0csgs7w.js}

@@ -1,12 +1,12 @@
-import {
-  stableArtifactRoot
-} from "./chunk-wpckd35c.js";
 import {
   workflow
-} from "./chunk-bfkmzv9h.js";
+} from "./chunk-xtxcq5tj.js";
 import {
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
+import {
+  stableArtifactRoot
+} from "./chunk-rc5ec5xx.js";
 
 // dist/builtin/workflows/builtin/classify-and-act.ts
 import { Type as Type2 } from "typebox";
@@ -112,7 +112,7 @@ async function runClassifyAndAct(ctx) {
   const confidenceValue = typeof value === "object" && value !== null && "confidence" in value ? value.confidence : 0;
   const confidence = typeof confidenceValue === "number" && Number.isFinite(confidenceValue) ? Math.max(0, Math.min(1, confidenceValue)) : 0;
   const rationale = typeof value === "object" && value !== null && "rationale" in value ? String(value.rationale) : "Classifier did not provide a usable structured rationale.";
-  const exactCategory = ctx.inputs.categories.find((category2) => category2 === proposedCategory);
+  const exactCategory = ctx.inputs.categories.find((category) => category === proposedCategory);
   const needsFallback = exactCategory === undefined || confidence < ctx.inputs.confidence_threshold;
   const fallback = needsFallback ? await resolveFallbackCategory(ctx, exactCategory) : undefined;
   const category = fallback?.category ?? exactCategory;
@@ -140,7 +140,7 @@ async function runClassifyAndAct(ctx) {
 }
 
 // dist/builtin/workflows/builtin/classify-and-act.ts
-var DEFAULT_ACTION_CATEGORIES = ["analysis", "implementation", "research"];
+var DEFAULT_ACTION_CATEGORIES2 = ["analysis", "implementation", "research"];
 var classify_and_act_default = workflow({
   name: "classify-and-act",
   description: "Classify a task with structured confidence, route deterministically to an isolated category action, and ask for human selection when confidence is low.",
@@ -150,7 +150,7 @@ var classify_and_act_default = workflow({
     categories: Type2.Array(Type2.String({ minLength: 1 }), {
       minItems: 1,
       maxItems: 8,
-      default: [...DEFAULT_ACTION_CATEGORIES],
+      default: [...DEFAULT_ACTION_CATEGORIES2],
       description: "Ordered action categories available to the classifier and fallback chooser."
     }),
     confidence_threshold: Type2.Number({
@@ -169,7 +169,7 @@ var classify_and_act_default = workflow({
     action_path: Type2.String({ description: "Category action report artifact path." }),
     artifact_dir: Type2.String({ description: "Per-run artifact directory." })
   },
-  run: async (ctx) => await runClassifyAndAct(withSteeringPropagationContext(ctx))
+  run: async (ctx) => await runClassifyAndAct(withSteeringPropagationContext2(ctx))
 });
 
-export { DEFAULT_ACTION_CATEGORIES, classify_and_act_default };
+export { DEFAULT_ACTION_CATEGORIES2, classify_and_act_default };

package/dist/builtin/workflows/builtin/{chunk-79fjkcvh.js → chunk-hbnwv56f.js}

@@ -1,25 +1,25 @@
 import {
-  fold_usage
-} from "./chunk-7at6dnkr.js";
-import {
-  build_scoring_prompt,
-  scoring_prompt_reads,
-  warm_first_fan_out
-} from "./chunk-7430zyas.js";
+  workflow
+} from "./chunk-xtxcq5tj.js";
 import {
-  stableArtifactRoot
-} from "./chunk-wpckd35c.js";
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
 import {
-  VERIFICATION_SCALE,
+  parse_rubric,
   normalize_criteria,
-  parse_rubric
-} from "./chunk-cg9tmks0.js";
+  VERIFICATION_SCALE
+} from "./chunk-n1910xc4.js";
 import {
-  workflow
-} from "./chunk-bfkmzv9h.js";
+  scoring_prompt_reads,
+  build_scoring_prompt,
+  warm_first_fan_out
+} from "./chunk-p8jjwfg6.js";
 import {
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  fold_usage
+} from "./chunk-tgt0s5e5.js";
+import {
+  stableArtifactRoot
+} from "./chunk-rc5ec5xx.js";
 
 // dist/builtin/workflows/builtin/tournament.ts
 import { Type as Type2 } from "typebox";
@@ -596,7 +596,7 @@ var tournament_default = workflow({
     seed: Type2.Integer({ description: "Seed used for the deterministic comparison schedule." }),
     artifact_dir: Type2.String({ description: "Run-specific directory containing tournament artifacts." })
   },
-  run: async (ctx) => await runTournament(withSteeringPropagationContext(ctx))
+  run: async (ctx) => await runTournament(withSteeringPropagationContext2(ctx))
 });
 
 export { tournament_default };

package/dist/builtin/workflows/builtin/{chunk-ed4jfn5t.js → chunk-mygmtz6e.js}

@@ -1,48 +1,48 @@
-import {
-  consolidateFindingsBatch,
-  convergence_escalation_evidence,
-  ensureWorkflowArtifactDirectory,
-  finalActionRemaining,
-  parseFailureDiagnostics,
-  record_convergence,
-  reverify_consolidated_batch,
-  reviewerFailureText,
-  summarizeReviewConvergence,
-  workflowArtifactDirectoryPath
-} from "./chunk-b0v2xab1.js";
-import {
-  fold_usage
-} from "./chunk-7at6dnkr.js";
-import {
-  VERIFICATION_SCALE
-} from "./chunk-cg9tmks0.js";
 import {
   workflow
-} from "./chunk-bfkmzv9h.js";
+} from "./chunk-xtxcq5tj.js";
 import {
-  ACCEPTANCE_MATRIX_CONTRACT,
-  CODE_QUALITY_VERIFICATION_GUIDANCE,
-  CONTRACT_FIDELITY_AUDIT,
+  keepContext,
+  REVIEWER_CALIBRATION_RULES,
+  WORKER_PREFLIGHT_CONTRACT,
   E2E_VERIFICATION_GUIDANCE,
-  EVIDENCE_CLOSURE_POLICY,
-  FINDINGS_CONSOLIDATION_CONTRACT,
-  LITERAL_OBJECTIVE_CONTRACT,
+  CODE_QUALITY_VERIFICATION_GUIDANCE,
   MEDIA_PUBLICATION_GUIDANCE,
-  REGRESSION_EVIDENCE_CONTRACT,
   REPO_INTENT_MINING_GUIDANCE,
-  REVIEWER_CALIBRATION_RULES,
-  REVIEWER_INDEPENDENT_VERIFICATION_CONTRACT,
-  REVIEWER_INTERCOM_COORDINATION_PROTOCOL,
-  REVIEWER_OVERIMPLEMENTATION_GUARD,
+  renderE2eQaVideoReviewGuidance,
+  LITERAL_OBJECTIVE_CONTRACT,
   REVIEWER_SPEC_VS_OBJECTIVE_GUARD,
-  REVIEW_CODE_DELTA_CONTRACT,
+  REVIEWER_OVERIMPLEMENTATION_GUARD,
+  ACCEPTANCE_MATRIX_CONTRACT,
+  CONTRACT_FIDELITY_AUDIT,
+  REVIEWER_INTERCOM_COORDINATION_PROTOCOL,
+  REVIEWER_INDEPENDENT_VERIFICATION_CONTRACT,
+  REGRESSION_EVIDENCE_CONTRACT,
+  FINDINGS_CONSOLIDATION_CONTRACT,
   SCOPE_DISCIPLINE_CONTRACT,
-  WORKER_PREFLIGHT_CONTRACT,
+  EVIDENCE_CLOSURE_POLICY,
   WORKTREE_DISCIPLINE_CONTRACT,
-  keepContext,
-  renderE2eQaVideoReviewGuidance,
-  withSteeringPropagationContext
-} from "./chunk-zbekjqew.js";
+  REVIEW_CODE_DELTA_CONTRACT,
+  withSteeringPropagationContext2
+} from "./chunk-a2kdsqn8.js";
+import {
+  workflowArtifactDirectoryPath,
+  ensureWorkflowArtifactDirectory,
+  record_convergence,
+  convergence_escalation_evidence,
+  finalActionRemaining,
+  consolidateFindingsBatch,
+  parseFailureDiagnostics,
+  reviewerFailureText,
+  summarizeReviewConvergence,
+  reverify_consolidated_batch
+} from "./chunk-zggjyg5z.js";
+import {
+  VERIFICATION_SCALE
+} from "./chunk-n1910xc4.js";
+import {
+  fold_usage
+} from "./chunk-tgt0s5e5.js";
 
 // dist/builtin/workflows/builtin/goal.ts
 import { Type as Type2 } from "typebox";
@@ -113,10 +113,10 @@ var reviewDecisionSchema = Type.Object({
 
 // dist/builtin/workflows/builtin/goal-models.ts
 var orchestratorModelConfig = {
-  model: "openai-codex/gpt-6-astra:high",
+  model: "openai-codex/gpt-6-astra:medium",
   fallbackModels: [
-    "github-copilot/gpt-6-astra:high",
-    "openai/gpt-6-astra:high",
+    "github-copilot/gpt-6-astra:medium",
+    "openai/gpt-6-astra:medium",
     "anthropic/claude-fable-5-1:high",
     "github-copilot/claude-fable-5-1:high",
     "anthropic/claude-opus-5:high",
@@ -142,7 +142,7 @@ var orchestratorModelConfig = {
     "zai-coding-cn/glm-5.3-flash:high",
     "baseten/zai-org/GLM-5.3:high",
     "baseten/zai-org/GLM-5.3-Flash:high",
-    "openrouter/openai/gpt-6-astra:high",
+    "openrouter/openai/gpt-6-astra:medium",
     "openrouter/anthropic/claude-fable-5-1:high",
     "openrouter/anthropic/claude-opus-5:high",
     "openrouter/openai/gpt-5.6-sol:xhigh",
@@ -151,7 +151,7 @@ var orchestratorModelConfig = {
     "openrouter/sakana/fugu-ultra:high",
     "openrouter/openai/gpt-5.5:xhigh",
     "openrouter/anthropic/claude-opus-4-8:high",
-    "openrouter/x-ai/grok-4.6",
+    "openrouter/x-ai/grok-4.6:xhigh",
     "openrouter/z-ai/glm-5.3:high",
     "openrouter/z-ai/glm-5.3-flash:high"
   ],
@@ -196,7 +196,7 @@ var reviewerModelConfig = {
     "openrouter/sakana/fugu-ultra:high",
     "openrouter/openai/gpt-5.5:xhigh",
     "openrouter/anthropic/claude-opus-4-8:high",
-    "openrouter/x-ai/grok-4.6",
+    "openrouter/x-ai/grok-4.6:xhigh",
     "openrouter/z-ai/glm-5.3:high",
     "openrouter/z-ai/glm-5.3-flash:high"
   ],
@@ -1227,7 +1227,7 @@ var goal_default = workflow({
     baseBranch: "base_branch"
   },
   run: async (ctx) => {
-    const workflowCtx = withSteeringPropagationContext(ctx);
+    const workflowCtx = withSteeringPropagationContext2(ctx);
     const workflowStartCwd = workflowCtx.cwd ?? process.cwd();
     const createPr = workflowCtx.inputs.create_pr === true;
     return await runGoalWorkflow(workflowCtx, { createPr, workflowStartCwd });