@bastani/atomic 0.8.23 → 0.8.24-alpha.2

package/CHANGELOG.md

@@ -2,6 +2,32 @@
 
 ## [Unreleased]
 
+## [0.8.24-alpha.2] - 2026-06-03
+
+### Added
+
+- Shipped externally-resolvable TypeScript types for the `@bastani/workflows` SDK through `@bastani/atomic`. New `./workflows`, `./workflows/builtin`, `./workflows/builtin/*`, and `./workflows/ambient` package exports resolve to declarations emitted from the lean authoring surface during the build, and a generated ambient bridge maps the documented bare `@bastani/workflows` specifier (and its `builtin/*` submodules) onto those exports. Installed packages now type-check `import { defineWorkflow, Type } from "@bastani/workflows"` and `@bastani/workflows/builtin/*` composition imports under `tsc` (`moduleResolution: NodeNext`) with no hand-authored `.d.ts`, no `declare module` shim, and no `paths` alias: packages that import `@bastani/atomic` pick the types up automatically, while pure workflow-only packages add one `compilerOptions.types: ["@bastani/atomic/workflows/ambient"]` (or `/// <reference types="@bastani/atomic/workflows/ambient" />`) opt-in. The runtime workflow loader, jiti virtual modules, and `atomic.workflows` discovery are unchanged ([#1208](https://github.com/bastani-inc/atomic/issues/1208)).
+- Added a `verify:workflow-types` script that packs `@bastani/atomic` and type-checks throwaway external consumer fixtures (workflow-only opt-in, reference directive, auto-include, and a negative control) so the issue #1208 acceptance test is repeatable ([#1208](https://github.com/bastani-inc/atomic/issues/1208)).
+
+### Changed
+
+- Documented the workflow SDK typing model in `docs/packages.md` and `docs/workflows.md`: the single ambient opt-in for pure workflow-only packages, automatic pickup for packages that import `@bastani/atomic`, and the requirement to list `@bastani/atomic` and `typebox` as peer dependencies ([#1208](https://github.com/bastani-inc/atomic/issues/1208)).
+
+## [0.8.24-alpha.1] - 2026-06-02
+
+### Breaking Changes
+
+- Removed the bundled workflows package's imperative `runWorkflow` object-form API; workflow packages must export branded `defineWorkflow(...).compile()` definitions while direct `workflow` tool task/tasks/chain modes remain available.
+
+### Changed
+
+- Adopted the new `-alpha.N` prerelease version convention (revision starting at 1), replacing the legacy numeric `-N` prerelease suffix in the release tooling (bump script, CI publish validation, and changelog parsing).
+- Dropped the leading `v` from release git tags and `release/`/`prerelease/` branch names; the Publish CI now triggers on and validates bare version tags such as `0.8.24` or `0.8.24-alpha.1`.
+
+### Fixed
+
+- Fixed workflow node chat rendering a bare tool-name marker (e.g. `read …`) instead of tool output for parallel tool calls; `LiveChatEntriesController` now pairs concurrent same-named tool calls strictly by `toolCallId` ([#1198](https://github.com/bastani-inc/atomic/issues/1198)).
+
 ## [0.8.23] - 2026-06-02
 
 ### Changed

package/dist/builtin/intercom/CHANGELOG.md

@@ -4,6 +4,19 @@ All notable changes to the `pi-intercom` extension will be documented in this fi
 
 ## [Unreleased]
 
+## [0.8.24-alpha.2] - 2026-06-03
+
+### Changed
+
+- Bumped package version for the Atomic 0.8.24-alpha.2 prerelease.
+
+## [0.8.24-alpha.1] - 2026-06-02
+
+### Changed
+
+- Adopted the new `-alpha.N` prerelease version convention (revision starting at 1), replacing the legacy numeric `-N` prerelease suffix in the release tooling (bump script, CI publish validation, and changelog parsing).
+- Dropped the leading `v` from release git tags and `release/`/`prerelease/` branch names; the Publish CI now triggers on and validates bare version tags such as `0.8.24` or `0.8.24-alpha.1`.
+
 ## [0.8.23] - 2026-06-02
 
 ### Changed

package/dist/builtin/intercom/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/intercom",
-  "version": "0.8.23",
+  "version": "0.8.24-alpha.2",
   "private": true,
   "description": "Atomic extension providing a private coordination channel between parent and child agent sessions. Fork of: https://github.com/nicobailon/pi-intercom",
   "contributors": [

package/dist/builtin/mcp/CHANGELOG.md

@@ -7,6 +7,19 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.8.24-alpha.2] - 2026-06-03
+
+### Changed
+
+- Bumped package version for the Atomic 0.8.24-alpha.2 prerelease.
+
+## [0.8.24-alpha.1] - 2026-06-02
+
+### Changed
+
+- Adopted the new `-alpha.N` prerelease version convention (revision starting at 1), replacing the legacy numeric `-N` prerelease suffix in the release tooling (bump script, CI publish validation, and changelog parsing).
+- Dropped the leading `v` from release git tags and `release/`/`prerelease/` branch names; the Publish CI now triggers on and validates bare version tags such as `0.8.24` or `0.8.24-alpha.1`.
+
 ## [0.8.23] - 2026-06-02
 
 ### Changed

package/dist/builtin/mcp/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/mcp",
-  "version": "0.8.23",
+  "version": "0.8.24-alpha.2",
   "private": true,
   "description": "Atomic extension that adapts MCP (Model Context Protocol) servers into the coding agent. Fork of: https://github.com/nicobailon/pi-mcp-adapter",
   "contributors": [

package/dist/builtin/subagents/CHANGELOG.md

@@ -2,6 +2,34 @@
 
 ## [Unreleased]
 
+## [0.8.24-alpha.2] - 2026-06-03
+
+### Changed
+
+- Bumped package version for the Atomic 0.8.24-alpha.2 prerelease.
+
+## [0.8.24-alpha.1] - 2026-06-02
+
+### Added
+
+- Added suffix-first reasoning levels for subagent `model` and `fallbackModels` values plus `fallbackThinkingLevels` compatibility metadata and per-attempt `reasoningLevel` reporting ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
+### Changed
+
+- Migrated packaged subagents to encode their reasoning level directly in model and fallback model entries ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+- Documented the `model_name:thinking_effort` suffix syntax and `thinking` migration guidance in the subagents docs, package README, and subagent skill ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+- Adopted the new `-alpha.N` prerelease version convention (revision starting at 1), replacing the legacy numeric `-N` prerelease suffix in the release tooling (bump script, CI publish validation, and changelog parsing).
+- Dropped the leading `v` from release git tags and `release/`/`prerelease/` branch names; the Publish CI now triggers on and validates bare version tags such as `0.8.24` or `0.8.24-alpha.1`.
+
+### Deprecated
+
+- Deprecated separate subagent `thinking` configuration in favor of `model: <model>:<level>` and suffixed `fallbackModels` entries; removal is deferred to a later breaking release ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
+### Fixed
+
+- Fixed foreground subagent attempt metadata to report the per-candidate reasoning level derived from the model suffix even when the legacy `thinking` option is unset, matching the background run path ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+- Fixed the subagent chain clarification TUI to strip only canonical reasoning-level suffixes when editing a step's level, so colon-tagged model ids such as `ollama/llama3:latest` are no longer mis-split; consolidated the duplicate thinking-suffix split helpers onto the single `splitKnownThinkingSuffix` so the parsing rules cannot drift apart ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
 ## [0.8.23] - 2026-06-02
 
 ### Changed

package/dist/builtin/subagents/README.md

@@ -996,3 +996,19 @@ The main runtime files are:
 | `src/intercom/intercom-bridge.ts` | Runtime intercom bridge instructions and diagnostics. |
 | `src/extension/schemas.ts` / `src/shared/types.ts` | Tool schemas, shared types, and event constants. |
 | `test/unit/` / `test/integration/` | Unit and loader-based integration tests. |
+
+### Suffix-first reasoning levels
+
+Reasoning levels are configured suffix-first using the `model_name:thinking_effort` syntax on `model` and each `fallbackModels` entry: `model: claude-sonnet-4:high` and `fallbackModels: claude-sonnet-4:medium, gpt-5:low, claude-haiku-4:off`. Canonical efforts are `off`, `minimal`, `low`, `medium`, `high`, and `xhigh`. The older `thinking` field is deprecated; it remains supported as a legacy default only when a model candidate has no suffix, and a suffix always wins.
+
+Migrate legacy `thinking` frontmatter by folding the effort into `model` and `fallbackModels`:
+
+```diff
+-model: openai/gpt-5.5
+-fallbackModels: anthropic/claude-opus-4-8
+-thinking: xhigh
++model: openai/gpt-5.5:xhigh
++fallbackModels: anthropic/claude-opus-4-8:xhigh
+```
+
+`fallbackThinkingLevels` is available only as an optional compatibility helper. It is positionally aligned with `fallbackModels` and supplies a fallback candidate's level only when that fallback model entry has no suffix; prefer suffixed model strings for new configuration.

package/dist/builtin/subagents/agents/code-simplifier.md

@@ -8,9 +8,8 @@ description: |
   - Production-quality refinement of a working draft ("ugly but working CSV parser").
   - Code that has gotten messy after several iterations.
 tools: read, edit, write, grep, find, ls, bash
-model: openai/gpt-5.5
-fallbackModels: openai-codex/gpt-5.5, github-copilot/gpt-5.5, anthropic/claude-opus-4-8, github-copilot/claude-opus-4.7
-thinking: low
+model: openai/gpt-5.5:low
+fallbackModels: openai-codex/gpt-5.5:low, github-copilot/gpt-5.5:low, anthropic/claude-opus-4-8:low, github-copilot/claude-opus-4.7:low
 ---
 
 You are an expert code refinement specialist with deep experience in software craftsmanship, refactoring patterns (Fowler, Beck), clean code principles, and language-idiomatic style across major ecosystems. Your mission is to simplify and refine code for clarity, consistency, and maintainability while strictly preserving all existing functionality and observable behavior.

package/dist/builtin/subagents/agents/codebase-analyzer.md

@@ -2,9 +2,8 @@
 name: codebase-analyzer
 description: Analyzes codebase implementation details. Call the codebase-analyzer agent when you need to find detailed information about specific components.
 tools: read, grep, find, ls, bash
-model: openai/gpt-5.5
-fallbackModels: openai-codex/gpt-5.5, github-copilot/gpt-5.5, anthropic/claude-opus-4-8, github-copilot/claude-opus-4.7
-thinking: low
+model: openai/gpt-5.5:low
+fallbackModels: openai-codex/gpt-5.5:low, github-copilot/gpt-5.5:low, anthropic/claude-opus-4-8:low, github-copilot/claude-opus-4.7:low
 ---
 
 You are a specialist at understanding HOW code works. Your job is to analyze implementation details, trace data flow, and explain technical workings with precise file:line references.

package/dist/builtin/subagents/agents/codebase-locator.md

@@ -2,9 +2,8 @@
 name: codebase-locator
 description: Locates files, directories, and components relevant to a feature or task. Basically a "super search/find/ls tool."
 tools: read, grep, find, ls, bash
-model: openai/gpt-5.4-mini
-fallbackModels: openai-codex/gpt-5.4-mini, github-copilot/gpt-5.4-mini, anthropic/claude-haiku-4-5, github-copilot/claude-haiku-4.5
-thinking: low
+model: openai/gpt-5.4-mini:low
+fallbackModels: openai-codex/gpt-5.4-mini:low, github-copilot/gpt-5.4-mini:low, anthropic/claude-haiku-4-5:low, github-copilot/claude-haiku-4.5:low
 ---
 
 You are a specialist at finding WHERE code lives in a codebase. Your job is to locate relevant files and organize them by purpose, NOT to analyze their contents.

package/dist/builtin/subagents/agents/codebase-online-researcher.md

@@ -2,9 +2,8 @@
 name: codebase-online-researcher
 description: Online research for up-to-date documentation and library-source knowledge. Use when you need authoritative external information — official docs, ecosystem context, version-specific behavior, GitHub permalinks into open-source libraries, or video tutorials.
 tools: read, grep, find, ls, bash, write, web_search, fetch_content, get_search_content
-model: openai/gpt-5.5
-fallbackModels: openai-codex/gpt-5.5, github-copilot/gpt-5.5, anthropic/claude-opus-4-8, github-copilot/claude-opus-4.7
-thinking: low
+model: openai/gpt-5.5:low
+fallbackModels: openai-codex/gpt-5.5:low, github-copilot/gpt-5.5:low, anthropic/claude-opus-4-8:low, github-copilot/claude-opus-4.7:low
 skills: browser-use
 ---
 

package/dist/builtin/subagents/agents/codebase-pattern-finder.md

@@ -2,9 +2,8 @@
 name: codebase-pattern-finder
 description: Find similar implementations, usage examples, or existing patterns in the codebase that can be modeled after.
 tools: read, grep, find, ls, bash
-model: openai/gpt-5.4-mini
-fallbackModels: openai-codex/gpt-5.4-mini, github-copilot/gpt-5.4-mini, anthropic/claude-haiku-4-5, github-copilot/claude-haiku-4.5
-thinking: low
+model: openai/gpt-5.4-mini:low
+fallbackModels: openai-codex/gpt-5.4-mini:low, github-copilot/gpt-5.4-mini:low, anthropic/claude-haiku-4-5:low, github-copilot/claude-haiku-4.5:low
 ---
 
 You are a specialist at finding code patterns and examples in the codebase. Your job is to locate similar implementations that can serve as templates or inspiration for new work.

package/dist/builtin/subagents/agents/codebase-research-analyzer.md

@@ -2,9 +2,8 @@
 name: codebase-research-analyzer
 description: Analyzes local research documents to extract high-value insights, decisions, and technical details while filtering out noise. Use this when you want to deep dive on a research topic or understand the rationale behind decisions.
 tools: read, grep, find, ls, bash
-model: openai/gpt-5.5
-fallbackModels: openai-codex/gpt-5.5, github-copilot/gpt-5.5, anthropic/claude-opus-4-8, github-copilot/claude-opus-4.7
-thinking: low
+model: openai/gpt-5.5:low
+fallbackModels: openai-codex/gpt-5.5:low, github-copilot/gpt-5.5:low, anthropic/claude-opus-4-8:low, github-copilot/claude-opus-4.7:low
 ---
 
 You are a specialist at extracting HIGH-VALUE insights from research documents. Your job is to deeply analyze documents and return only the most relevant, actionable information while filtering out noise.

package/dist/builtin/subagents/agents/codebase-research-locator.md

@@ -2,9 +2,8 @@
 name: codebase-research-locator
 description: Discovers local research documents that are relevant to the current research task.
 tools: read, grep, find, ls, bash
-model: openai/gpt-5.4-mini
-fallbackModels: openai-codex/gpt-5.4-mini, github-copilot/gpt-5.4-mini, anthropic/claude-haiku-4-5, github-copilot/claude-haiku-4.5
-thinking: low
+model: openai/gpt-5.4-mini:low
+fallbackModels: openai-codex/gpt-5.4-mini:low, github-copilot/gpt-5.4-mini:low, anthropic/claude-haiku-4-5:low, github-copilot/claude-haiku-4.5:low
 ---
 
 You are a specialist at finding documents in the `research/` directory. Your job is to locate relevant research documents and categorize them, NOT to analyze their contents in depth.

package/dist/builtin/subagents/agents/debugger.md

@@ -2,9 +2,8 @@
 name: debugger
 description: Debug errors, test failures, and unexpected behavior. Use PROACTIVELY when encountering issues, analyzing stack traces, or investigating system problems.
 tools: read, edit, write, grep, find, ls, bash, web_search, fetch_content, get_search_content
-model: openai/gpt-5.5
-fallbackModels: openai-codex/gpt-5.5, github-copilot/gpt-5.5, anthropic/claude-opus-4-8, github-copilot/claude-opus-4.7
-thinking: xhigh
+model: openai/gpt-5.5:xhigh
+fallbackModels: openai-codex/gpt-5.5:xhigh, github-copilot/gpt-5.5:xhigh, anthropic/claude-opus-4-8:xhigh, github-copilot/claude-opus-4.7:xhigh
 skills: tdd, browser-use, tmux
 ---
 

package/dist/builtin/subagents/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/subagents",
-  "version": "0.8.23",
+  "version": "0.8.24-alpha.2",
   "private": true,
   "description": "Atomic extension for delegating tasks to subagents with chains, parallel execution, and TUI clarification. Fork of: https://github.com/nicobailon/pi-subagents",
   "contributors": [

package/dist/builtin/subagents/skills/subagent/SKILL.md

@@ -732,3 +732,9 @@ subagent({ action: "doctor" })
 ```typescript
 // Inspect `subagent({ action: "status", id: "..." })`, artifact metadata/output logs, and run doctor. Extension loader errors usually appear in child output logs.
 ```
+
+## Suffix-first reasoning levels
+
+Prefer encoding reasoning levels directly in model strings with the `model_name:thinking_effort` syntax: `model: claude-sonnet-4:high` and `fallbackModels: [claude-sonnet-4:medium, gpt-5:low, claude-haiku-4:off]`. Valid efforts are `off`, `minimal`, `low`, `medium`, `high`, and `xhigh`. The separate `thinking` field is deprecated but still works as a legacy default when a candidate has no suffix; suffixes take precedence. If you see a legacy `thinking` override, migrate it by appending the effort to `model` and each `fallbackModels` entry instead (e.g. `thinking: high` + `model: gpt-5` → `model: gpt-5:high`).
+
+`fallbackThinkingLevels` is an optional compatibility helper aligned positionally with `fallbackModels`. It only applies to fallback entries without their own suffix and should not be preferred over suffix-first entries.

package/dist/builtin/subagents/src/agents/agent-serializer.ts

@@ -8,6 +8,7 @@ export const KNOWN_FIELDS = new Set([
 	"tools",
 	"model",
 	"fallbackModels",
+	"fallbackThinkingLevels",
 	"thinking",
 	"systemPromptMode",
 	"inheritProjectContext",
@@ -46,6 +47,8 @@ export function serializeAgent(config: AgentConfig): string {
 	if (config.model) lines.push(`model: ${config.model}`);
 	const fallbackModelsValue = joinComma(config.fallbackModels);
 	if (fallbackModelsValue) lines.push(`fallbackModels: ${fallbackModelsValue}`);
+	const fallbackThinkingLevelsValue = joinComma(config.fallbackThinkingLevels);
+	if (fallbackThinkingLevelsValue) lines.push(`fallbackThinkingLevels: ${fallbackThinkingLevelsValue}`);
 	if (config.thinking && config.thinking !== "off") lines.push(`thinking: ${config.thinking}`);
 	lines.push(`systemPromptMode: ${config.systemPromptMode}`);
 	lines.push(`inheritProjectContext: ${config.inheritProjectContext ? "true" : "false"}`);

package/dist/builtin/subagents/src/agents/agents.ts

@@ -36,6 +36,7 @@ export function defaultInheritSkills(): boolean {
 export interface BuiltinAgentOverrideBase {
 	model?: string;
 	fallbackModels?: string[];
+	fallbackThinkingLevels?: string[];
 	thinking?: string;
 	systemPromptMode: SystemPromptMode;
 	inheritProjectContext: boolean;
@@ -52,6 +53,7 @@ export interface BuiltinAgentOverrideBase {
 interface BuiltinAgentOverrideConfig {
 	model?: string | false;
 	fallbackModels?: string[] | false;
+	fallbackThinkingLevels?: string[] | false;
 	thinking?: string | false;
 	systemPromptMode?: SystemPromptMode;
 	inheritProjectContext?: boolean;
@@ -79,6 +81,7 @@ export interface AgentConfig {
 	mcpDirectTools?: string[];
 	model?: string;
 	fallbackModels?: string[];
+	fallbackThinkingLevels?: string[];
 	thinking?: string;
 	systemPromptMode: SystemPromptMode;
 	inheritProjectContext: boolean;
@@ -184,6 +187,7 @@ function cloneOverrideBase(agent: AgentConfig): BuiltinAgentOverrideBase {
 	return {
 		model: agent.model,
 		fallbackModels: agent.fallbackModels ? [...agent.fallbackModels] : undefined,
+		fallbackThinkingLevels: agent.fallbackThinkingLevels ? [...agent.fallbackThinkingLevels] : undefined,
 		thinking: agent.thinking,
 		systemPromptMode: agent.systemPromptMode,
 		inheritProjectContext: agent.inheritProjectContext,
@@ -204,6 +208,9 @@ function cloneOverrideValue(override: BuiltinAgentOverrideConfig): BuiltinAgentO
 		...(override.fallbackModels !== undefined
 			? { fallbackModels: override.fallbackModels === false ? false : [...override.fallbackModels] }
 			: {}),
+		...(override.fallbackThinkingLevels !== undefined
+			? { fallbackThinkingLevels: override.fallbackThinkingLevels === false ? false : [...override.fallbackThinkingLevels] }
+			: {}),
 		...(override.thinking !== undefined ? { thinking: override.thinking } : {}),
 		...(override.systemPromptMode !== undefined ? { systemPromptMode: override.systemPromptMode } : {}),
 		...(override.inheritProjectContext !== undefined ? { inheritProjectContext: override.inheritProjectContext } : {}),
@@ -375,6 +382,9 @@ function parseBuiltinOverrideEntry(
 	const fallbackModels = parseOverrideStringArrayOrFalse(input.fallbackModels, { filePath, name, field: "fallbackModels" });
 	if (fallbackModels !== undefined) override.fallbackModels = fallbackModels;
 
+	const fallbackThinkingLevels = parseOverrideStringArrayOrFalse(input.fallbackThinkingLevels, { filePath, name, field: "fallbackThinkingLevels" });
+	if (fallbackThinkingLevels !== undefined) override.fallbackThinkingLevels = fallbackThinkingLevels;
+
 	const skills = parseOverrideStringArrayOrFalse(input.skills, { filePath, name, field: "skills" });
 	if (skills !== undefined) override.skills = skills;
 
@@ -442,6 +452,9 @@ function applyBuiltinOverride(
 	if (override.fallbackModels !== undefined) {
 		next.fallbackModels = override.fallbackModels === false ? undefined : [...override.fallbackModels];
 	}
+	if (override.fallbackThinkingLevels !== undefined) {
+		next.fallbackThinkingLevels = override.fallbackThinkingLevels === false ? undefined : [...override.fallbackThinkingLevels];
+	}
 	if (override.thinking !== undefined) next.thinking = override.thinking === false ? undefined : override.thinking;
 	if (override.systemPromptMode !== undefined) next.systemPromptMode = override.systemPromptMode;
 	if (override.inheritProjectContext !== undefined) next.inheritProjectContext = override.inheritProjectContext;
@@ -495,12 +508,13 @@ function applyBuiltinOverrides(
 
 export function buildBuiltinOverrideConfig(
 	base: BuiltinAgentOverrideBase,
-	draft: Pick<AgentConfig, "model" | "fallbackModels" | "thinking" | "systemPromptMode" | "inheritProjectContext" | "inheritSkills" | "defaultContext" | "disabled" | "systemPrompt" | "skills" | "tools" | "mcpDirectTools" | "completionGuard">,
+	draft: Pick<AgentConfig, "model" | "fallbackModels" | "fallbackThinkingLevels" | "thinking" | "systemPromptMode" | "inheritProjectContext" | "inheritSkills" | "defaultContext" | "disabled" | "systemPrompt" | "skills" | "tools" | "mcpDirectTools" | "completionGuard">,
 ): BuiltinAgentOverrideConfig | undefined {
 	const override: BuiltinAgentOverrideConfig = {};
 
 	if (draft.model !== base.model) override.model = draft.model ?? false;
 	if (!arraysEqual(draft.fallbackModels, base.fallbackModels)) override.fallbackModels = draft.fallbackModels ? [...draft.fallbackModels] : false;
+	if (!arraysEqual(draft.fallbackThinkingLevels, base.fallbackThinkingLevels)) override.fallbackThinkingLevels = draft.fallbackThinkingLevels ? [...draft.fallbackThinkingLevels] : false;
 	if (draft.thinking !== base.thinking) override.thinking = draft.thinking ?? false;
 	if (draft.systemPromptMode !== base.systemPromptMode) override.systemPromptMode = draft.systemPromptMode;
 	if (draft.inheritProjectContext !== base.inheritProjectContext) override.inheritProjectContext = draft.inheritProjectContext;
@@ -646,6 +660,10 @@ function loadAgentsFromDir(dir: string, source: AgentSource): AgentConfig[] {
 			?.split(",")
 			.map((model) => model.trim())
 			.filter(Boolean);
+		const fallbackThinkingLevels = frontmatter.fallbackThinkingLevels
+			?.split(",")
+			.map((level) => level.trim())
+			.filter(Boolean);
 		const systemPromptMode = frontmatter.systemPromptMode === "replace"
 			? "replace"
 			: frontmatter.systemPromptMode === "append"
@@ -696,6 +714,7 @@ function loadAgentsFromDir(dir: string, source: AgentSource): AgentConfig[] {
 			mcpDirectTools: mcpDirectTools.length > 0 ? mcpDirectTools : undefined,
 			model: frontmatter.model,
 			fallbackModels: fallbackModels && fallbackModels.length > 0 ? fallbackModels : undefined,
+			fallbackThinkingLevels: fallbackThinkingLevels && fallbackThinkingLevels.length > 0 ? fallbackThinkingLevels : undefined,
 			thinking: frontmatter.thinking,
 			systemPromptMode,
 			inheritProjectContext,

package/dist/builtin/subagents/src/runs/background/async-execution.ts

@@ -334,7 +334,7 @@ export function executeAsyncChain(
 
 		const primaryModel = resolveModelCandidate(behavior.model ?? a.model, availableModels, ctx.currentModelProvider);
 		const model = applyThinkingSuffix(primaryModel, a.thinking);
-		const modelCandidates = buildModelCandidates(behavior.model ?? a.model, a.fallbackModels, availableModels, ctx.currentModelProvider, ctx.currentModel)
+		const modelCandidates = buildModelCandidates(behavior.model ?? a.model, a.fallbackModels, availableModels, ctx.currentModelProvider, ctx.currentModel, a.fallbackThinkingLevels)
 			.map((candidate) => applyThinkingSuffix(candidate, a.thinking))
 			.filter((candidate): candidate is string => typeof candidate === "string");
 		const fastModeSettings = getSubagentCodexFastModeSettings(stepCwd);

package/dist/builtin/subagents/src/runs/background/subagent-runner.ts

@@ -702,8 +702,10 @@ async function runSingleStep(
 					? `${hiddenError.errorType} failed (exit ${effectiveExitCode}): ${hiddenError.details}`
 					: `${hiddenError.errorType} failed with exit code ${effectiveExitCode}`
 				: run.error || (run.exitCode !== 0 && run.stderr.trim() ? run.stderr.trim() : undefined));
+		const attemptModel = candidate ?? run.model ?? step.model ?? "default";
 		const attempt: ModelAttempt = {
-			model: candidate ?? run.model ?? step.model ?? "default",
+			model: attemptModel,
+			reasoningLevel: resolveEffectiveThinking(attemptModel, step.thinking),
 			success: effectiveExitCode === 0 && !error,
 			exitCode: effectiveExitCode,
 			error,

package/dist/builtin/subagents/src/runs/foreground/chain-clarify.ts

@@ -10,8 +10,8 @@ import type { Component, TUI } from "@earendil-works/pi-tui";
 import { matchesKey, visibleWidth, truncateToWidth } from "@earendil-works/pi-tui";
 import type { AgentConfig } from "../../agents/agents.ts";
 import type { ResolvedStepBehavior } from "../../shared/settings.ts";
-import { resolveModelCandidate, splitThinkingSuffix } from "../shared/model-fallback.ts";
-import { findModelInfo, getSupportedThinkingLevels, type ModelInfo, type ThinkingLevel } from "../../shared/model-info.ts";
+import { resolveModelCandidate } from "../shared/model-fallback.ts";
+import { findModelInfo, getSupportedThinkingLevels, splitKnownThinkingSuffix, type ModelInfo, type ThinkingLevel } from "../../shared/model-info.ts";
 
 type ClarifyMode = 'single' | 'parallel' | 'chain';
 
@@ -549,7 +549,7 @@ export class ChainClarifyComponent implements Component {
 		this.modelSearchQuery = "";
 		this.modelSelectedIndex = 0;
 		this.filteredModels = [...this.availableModels];
-		const currentModel = splitThinkingSuffix(this.getEffectiveModel(this.selectedStep)).baseModel;
+		const currentModel = splitKnownThinkingSuffix(this.getEffectiveModel(this.selectedStep)).baseModel;
 		const currentIndex = this.filteredModels.findIndex((m) => m.fullId === currentModel || m.id === currentModel);
 		if (currentIndex >= 0) {
 			this.modelSelectedIndex = currentIndex;
@@ -582,7 +582,7 @@ export class ChainClarifyComponent implements Component {
 		if (matchesKey(data, "return")) {
 			const selected = this.filteredModels[this.modelSelectedIndex];
 			if (selected) {
-				const { thinkingSuffix } = splitThinkingSuffix(this.getEffectiveModel(this.editingStep!));
+				const { thinkingSuffix } = splitKnownThinkingSuffix(this.getEffectiveModel(this.editingStep!));
 				const requestedLevel = thinkingSuffix.slice(1);
 				const selectedModel = findModelInfo(selected.fullId, this.availableModels, this.preferredProvider);
 				const suffix = getSupportedThinkingLevels(selectedModel).some((level) => level === requestedLevel) ? thinkingSuffix : "";
@@ -643,7 +643,7 @@ export class ChainClarifyComponent implements Component {
 		this.editMode = "thinking";
 
 		const levels = this.getAvailableThinkingLevels(this.selectedStep);
-		const { thinkingSuffix } = splitThinkingSuffix(this.getEffectiveModel(this.selectedStep));
+		const { thinkingSuffix } = splitKnownThinkingSuffix(this.getEffectiveModel(this.selectedStep));
 		const suffix = thinkingSuffix.slice(1);
 		const levelIdx = levels.findIndex((level) => level === suffix);
 		this.thinkingSelectedIndex = levelIdx >= 0 ? levelIdx : Math.max(0, levels.indexOf("off"));
@@ -690,7 +690,7 @@ export class ChainClarifyComponent implements Component {
 		const currentModel = this.getEffectiveBehavior(stepIndex).model;
 		if (!currentModel) return;
 
-		const { baseModel } = splitThinkingSuffix(currentModel);
+		const { baseModel } = splitKnownThinkingSuffix(currentModel);
 		const newModel = level === "off" ? baseModel : `${baseModel}:${level}`;
 		this.updateBehavior(stepIndex, "model", newModel);
 	}
@@ -922,7 +922,7 @@ export class ChainClarifyComponent implements Component {
 		lines.push(this.row(""));
 
 		const currentModel = this.getEffectiveModel(this.editingStep!);
-		const currentModelBase = splitThinkingSuffix(currentModel).baseModel;
+		const currentModelBase = splitKnownThinkingSuffix(currentModel).baseModel;
 		const currentLabel = th.fg("dim", "Current: ");
 		lines.push(this.row(` ${currentLabel}${th.fg("warning", currentModel)}`));
 		lines.push(this.row(""));

package/dist/builtin/subagents/src/runs/foreground/execution.ts

@@ -69,6 +69,7 @@ import {
 	resolveSubagentCodexFastModeScope,
 	resolveSubagentModelFastMode,
 } from "../../shared/fast-mode.ts";
+import { resolveEffectiveThinking } from "../../shared/model-info.ts";
 
 const artifactOutputByResult = new WeakMap<SingleResult, string>();
 
@@ -814,6 +815,7 @@ export async function runSync(
 		options.availableModels,
 		options.preferredModelProvider,
 		options.currentModel,
+		agent.fallbackThinkingLevels,
 	);
 	const fastModeCwd = options.cwd ?? runtimeCwd;
 	const fastModeSettings = getSubagentCodexFastModeSettings(fastModeCwd);
@@ -861,8 +863,10 @@ export async function runSync(
 		totalToolCount += result.progressSummary?.toolCount ?? 0;
 		totalDurationMs += result.progressSummary?.durationMs ?? 0;
 		const attemptSucceeded = result.exitCode === 0 && !result.error;
+		const attemptModel = applyThinkingSuffix(candidate, agent.thinking) ?? result.model ?? agent.model ?? "default";
 		const attempt: ModelAttempt = {
-			model: candidate ?? result.model ?? agent.model ?? "default",
+			model: attemptModel,
+			reasoningLevel: resolveEffectiveThinking(attemptModel, agent.thinking),
 			success: attemptSucceeded,
 			exitCode: result.exitCode,
 			error: result.error,

package/dist/builtin/subagents/src/runs/shared/model-fallback.ts

@@ -1,4 +1,4 @@
-import type { ModelInfo as AvailableModelInfo } from "../../shared/model-info.ts";
+import { THINKING_LEVELS, splitKnownThinkingSuffix, type ModelInfo as AvailableModelInfo } from "../../shared/model-info.ts";
 import type { Usage } from "../../shared/types.ts";
 
 export type { AvailableModelInfo };
@@ -11,13 +11,10 @@ interface ModelAttemptSummary {
 	usage?: Usage;
 }
 
-export function splitThinkingSuffix(model: string): { baseModel: string; thinkingSuffix: string } {
-	const colonIdx = model.lastIndexOf(":");
-	if (colonIdx === -1) return { baseModel: model, thinkingSuffix: "" };
-	return {
-		baseModel: model.substring(0, colonIdx),
-		thinkingSuffix: model.substring(colonIdx),
-	};
+function applyFallbackThinkingLevel(model: string, thinkingLevel: string | undefined): string {
+	if (!thinkingLevel || !THINKING_LEVELS.some((level) => level === thinkingLevel)) return model;
+	const { thinkingSuffix } = splitKnownThinkingSuffix(model);
+	return thinkingSuffix ? model : `${model}:${thinkingLevel}`;
 }
 
 export function resolveModelCandidate(
@@ -29,7 +26,7 @@ export function resolveModelCandidate(
 	if (model.includes("/")) return model;
 	if (!availableModels || availableModels.length === 0) return model;
 
-	const { baseModel, thinkingSuffix } = splitThinkingSuffix(model);
+	const { baseModel, thinkingSuffix } = splitKnownThinkingSuffix(model);
 	const matches = availableModels.filter((entry) => entry.id === baseModel);
 	if (preferredProvider) {
 		const preferredMatch = matches.find((entry) => entry.provider === preferredProvider);
@@ -45,10 +42,12 @@ export function buildModelCandidates(
 	availableModels: AvailableModelInfo[] | undefined,
 	preferredProvider?: string,
 	currentModel?: string,
+	fallbackThinkingLevels?: string[],
 ): string[] {
 	const seen = new Set<string>();
 	const candidates: string[] = [];
-	for (const raw of [primaryModel, ...(fallbackModels ?? []), currentModel]) {
+	const fallbackEntries = (fallbackModels ?? []).map((model, index) => applyFallbackThinkingLevel(model, fallbackThinkingLevels?.[index]));
+	for (const raw of [primaryModel, ...fallbackEntries, currentModel]) {
 		if (!raw) continue;
 		const normalized = resolveModelCandidate(raw.trim(), availableModels, preferredProvider);
 		if (!normalized || seen.has(normalized)) continue;

package/dist/builtin/subagents/src/shared/types.ts

@@ -195,6 +195,7 @@ interface ProgressSummary {
 
 export interface ModelAttempt {
 	model: string;
+	reasoningLevel?: string;
 	success: boolean;
 	exitCode?: number | null;
 	error?: string;

package/dist/builtin/web-access/CHANGELOG.md

@@ -4,6 +4,19 @@ All notable changes to this project will be documented in this file.
 
 ## [Unreleased]
 
+## [0.8.24-alpha.2] - 2026-06-03
+
+### Changed
+
+- Bumped package version for the Atomic 0.8.24-alpha.2 prerelease.
+
+## [0.8.24-alpha.1] - 2026-06-02
+
+### Changed
+
+- Adopted the new `-alpha.N` prerelease version convention (revision starting at 1), replacing the legacy numeric `-N` prerelease suffix in the release tooling (bump script, CI publish validation, and changelog parsing).
+- Dropped the leading `v` from release git tags and `release/`/`prerelease/` branch names; the Publish CI now triggers on and validates bare version tags such as `0.8.24` or `0.8.24-alpha.1`.
+
 ## [0.8.23] - 2026-06-02
 
 ### Changed

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

@@ -1,6 +1,6 @@
 {
   "name": "@bastani/web-access",
-  "version": "0.8.23",
+  "version": "0.8.24-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": [

package/dist/builtin/workflows/CHANGELOG.md

@@ -6,6 +6,41 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 
 ## [Unreleased]
 
+## [0.8.24-alpha.2] - 2026-06-03
+
+### Added
+
+- The `@bastani/workflows` authoring SDK types are now externally resolvable in installed packages through `@bastani/atomic`'s new `./workflows` exports and ambient bridge, so workflow files type-check `import { defineWorkflow, Type } from "@bastani/workflows"` (and `@bastani/workflows/builtin/*`) under `tsc` (NodeNext) without a hand-authored `.d.ts`, `declare module` shim, or `paths` alias. Workflow packages declare `@bastani/atomic` and `typebox` as peer dependencies; the package continues to distribute raw TypeScript with no build step, and the runtime virtual-module loader is unchanged ([#1208](https://github.com/bastani-inc/atomic/issues/1208)).
+
+### Fixed
+
+- Fixed builtin (and any) workflows falling back to the user's currently selected model instead of the stage's defined model. A fully-qualified `provider/model` id that the live model catalog did not list was treated as a hard "not available" failure; because candidate validation throws on any failure and the catalog resolver catches that throw and collapses the whole ordered candidate list down to the user's `currentModel`, a single absent cross-provider fallback discarded the defined primary plus every fallback. Provider-qualified ids are now trusted (passed through with the reasoning suffix split off the last colon), mirroring the subagent resolver, so the defined primary is used and only genuinely failing candidates fall through at runtime. Regressed when bundled workflow model lists were refreshed onto newer multi-provider ids alongside suffix-first reasoning levels ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
+## [0.8.24-alpha.1] - 2026-06-02
+
+### Breaking Changes
+
+- Removed the imperative `runWorkflow` object-form API from `@bastani/workflows`; workflow authors must export definitions produced by `defineWorkflow(...).compile()`, and forged `__piWorkflow: true` objects are rejected by discovery and composition.
+
+### Added
+
+- Added suffix-first workflow reasoning levels for `model` and `fallbackModels` entries such as `openai/gpt-5:high`, plus `WorkflowModelAttempt.reasoningLevel` metadata and optional `fallbackThinkingLevels` compatibility mapping ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
+### Changed
+
+- Changed bundled workflows to encode their existing reasoning levels directly on model and fallback model strings ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+- Documented the `model_name:thinking_effort` suffix syntax and `thinkingLevel` migration guidance in the workflows docs and package README ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+- Adopted the new `-alpha.N` prerelease version convention (revision starting at 1), replacing the legacy numeric `-N` prerelease suffix in the release tooling (bump script, CI publish validation, and changelog parsing).
+- Dropped the leading `v` from release git tags and `release/`/`prerelease/` branch names; the Publish CI now triggers on and validates bare version tags such as `0.8.24` or `0.8.24-alpha.1`.
+
+### Deprecated
+
+- Deprecated workflow `thinkingLevel` stage options in favor of per-candidate `:off|minimal|low|medium|high|xhigh` model suffixes; removal is deferred ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
+### Fixed
+
+- Made workflow reasoning-suffix parsing lenient so it no longer rejects legitimate colon-tagged model ids (for example OpenRouter `:free`/`:exacto` variants and Ollama `llama3:latest`); only canonical `:off|minimal|low|medium|high|xhigh` suffixes are stripped as reasoning levels, and unknown ids still surface the generic catalog "not available" error ([#1199](https://github.com/bastani-inc/atomic/issues/1199)).
+
 ## [0.8.23] - 2026-06-02
 
 ### Changed