proseforge-agent 0.3.1__tar.gz → 0.3.2__tar.gz

{proseforge_agent-0.3.1/src/proseforge_agent.egg-info → proseforge_agent-0.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: proseforge-agent
-Version: 0.3.1
+Version: 0.3.2
 Summary: Agentic orchestration layer for long-form novel production on top of the ProseForge engine.
 Author: ProseForge Agent
 License: Apache-2.0
@@ -32,7 +32,7 @@ Dynamic: license-file
 <p align="center">
   <a href="README.zh-CN.md">简体中文</a> ·
   <a href="https://pypi.org/project/proseforge-agent/">PyPI</a> ·
-  <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.1">Release v0.3.1</a> ·
+  <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.2">Release v0.3.2</a> ·
   <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/pkgs/container/proseforge-agent">GHCR</a>
 </p>
 
@@ -59,7 +59,7 @@ image, or a background agent with gateway integrations.
 <tr><td><b>Review and revise</b></td><td>Run a chapter lifecycle from draft to review, rewrite, and accept, with quality gates and rollback-friendly artifacts.</td></tr>
 <tr><td><b>Use the model you want</b></td><td>Profiles for OpenAI, Anthropic, Gemini, xAI/Grok, DeepSeek, Qwen, GLM, Mimo, MiniMax, Doubao, plus an offline fake provider.</td></tr>
 <tr><td><b>Run it like a real tool</b></td><td>Full-screen TUI by default, explicit chat command, local API, Docker image, Telegram gateway, MCP stdio, local execution, cron, and telemetry.</td></tr>
-<tr><td><b>Published now</b></td><td><code>v0.3.1</code> is live on PyPI, GHCR, and GitHub Releases. Release gate: <code>1358 tests passing</code> — 35 integration (real subprocess / HTTP / SQLite), the rest offline fakes.</td></tr>
+<tr><td><b>Published now</b></td><td><code>v0.3.2</code> is the first-run UX release: one-line install upgrades the PyPI tool, repairs PATH priority, runs doctor, and opens the guided setup/TUI flow.</td></tr>
 </table>
 
 ---
@@ -85,6 +85,12 @@ curl -fsSL https://raw.githubusercontent.com/remacheybn408-boop/ProseForge-Agent
 iex (irm https://raw.githubusercontent.com/remacheybn408-boop/ProseForge-Agent/main/scripts/install.ps1)
 ```
 
+The one-line installer uses PyPI + `uv tool` by default. It installs or
+upgrades `proseforge-agent`, puts the manager bin directory first on PATH,
+runs `pf-agent doctor`, then starts `pf-agent setup --quick`. Use
+`--no-setup`, `--no-launch`, or `--non-interactive` on POSIX, and
+`-NoSetup`, `-NoLaunch`, or `-NonInteractive` on PowerShell, when scripting.
+
 ### Docker
 
 ```bash
@@ -99,7 +105,7 @@ want unreleased changes from `main`.
 
 ```bash
 pf-agent quickstart            # 60-second offline tour, no API keys needed
-pf-agent setup                 # guided wizard: providers, keys, workspace
+pf-agent setup --quick         # guided wizard: provider, API key, model, language, workspace
 pf-agent                       # full-screen terminal UI
 pf-agent tui                   # same TUI, with explicit flags
 pf-agent chat --message "draft an opening" --provider fake
@@ -109,13 +115,13 @@ pf-agent service serve         # local HTTP API on 127.0.0.1:8765
 
 The bare `pf-agent` command works like other coding agents' CLIs: the
 first run bootstraps a workspace, then opens the full-screen terminal UI.
-After `pf-agent setup` it talks to your configured default model. Without
+After `pf-agent setup --quick` it talks to your configured default model. Without
 setup, it runs against the offline fake provider so you can tour the CLI
 with no keys.
 
-Set provider API keys in your environment (or `.env`, see
-`.env.example`): `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, and friends.
-Every command accepts `--provider fake` for a no-key offline tour.
+The setup wizard writes API keys to `.pf-agent/.env`; `config.yaml` stores
+only provider metadata and `api_key_ref`. Every command accepts
+`--provider fake` for a no-key offline tour.
 
 ## Project Boundary
 
@@ -127,9 +133,9 @@ workflow state, conversational agent loops, and background event handling.
 
 ## Release Status
 
-`v0.3.1` was published on July 3, 2026 through PyPI Trusted Publishing.
-Release links: [PyPI](https://pypi.org/project/proseforge-agent/0.3.1/),
-[GitHub Release](https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.1),
+`v0.3.2` is prepared as the first-run UX patch release for July 3, 2026.
+Release links: [PyPI](https://pypi.org/project/proseforge-agent/0.3.2/),
+[GitHub Release](https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.2),
 [GHCR image](https://github.com/remacheybn408-boop/ProseForge-Agent/pkgs/container/proseforge-agent),
 and [`CHANGELOG.md`](CHANGELOG.md).
 
@@ -323,7 +329,7 @@ requests re-checked by downstream policy, plus research-ready
 `PyPIPublisher` — real `python -m build` + `twine upload` for TestPyPI /
 PyPI, gated by the Task 47 `PackageChecker` and duplicate-version refusal,
 with injectable runner + `dry_run` for offline tests; tokens never appear
-in the plan, report, or logs. The public `v0.3.1` release path uses PyPI
+in the plan, report, or logs. The public `v0.3.2` release path uses PyPI
 Trusted Publishing / OIDC, so no long-lived PyPI token is stored in the
 repository.
 

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/README.md

@@ -13,7 +13,7 @@
 <p align="center">
   <a href="README.zh-CN.md">简体中文</a> ·
   <a href="https://pypi.org/project/proseforge-agent/">PyPI</a> ·
-  <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.1">Release v0.3.1</a> ·
+  <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.2">Release v0.3.2</a> ·
   <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/pkgs/container/proseforge-agent">GHCR</a>
 </p>
 
@@ -40,7 +40,7 @@ image, or a background agent with gateway integrations.
 <tr><td><b>Review and revise</b></td><td>Run a chapter lifecycle from draft to review, rewrite, and accept, with quality gates and rollback-friendly artifacts.</td></tr>
 <tr><td><b>Use the model you want</b></td><td>Profiles for OpenAI, Anthropic, Gemini, xAI/Grok, DeepSeek, Qwen, GLM, Mimo, MiniMax, Doubao, plus an offline fake provider.</td></tr>
 <tr><td><b>Run it like a real tool</b></td><td>Full-screen TUI by default, explicit chat command, local API, Docker image, Telegram gateway, MCP stdio, local execution, cron, and telemetry.</td></tr>
-<tr><td><b>Published now</b></td><td><code>v0.3.1</code> is live on PyPI, GHCR, and GitHub Releases. Release gate: <code>1358 tests passing</code> — 35 integration (real subprocess / HTTP / SQLite), the rest offline fakes.</td></tr>
+<tr><td><b>Published now</b></td><td><code>v0.3.2</code> is the first-run UX release: one-line install upgrades the PyPI tool, repairs PATH priority, runs doctor, and opens the guided setup/TUI flow.</td></tr>
 </table>
 
 ---
@@ -66,6 +66,12 @@ curl -fsSL https://raw.githubusercontent.com/remacheybn408-boop/ProseForge-Agent
 iex (irm https://raw.githubusercontent.com/remacheybn408-boop/ProseForge-Agent/main/scripts/install.ps1)
 ```
 
+The one-line installer uses PyPI + `uv tool` by default. It installs or
+upgrades `proseforge-agent`, puts the manager bin directory first on PATH,
+runs `pf-agent doctor`, then starts `pf-agent setup --quick`. Use
+`--no-setup`, `--no-launch`, or `--non-interactive` on POSIX, and
+`-NoSetup`, `-NoLaunch`, or `-NonInteractive` on PowerShell, when scripting.
+
 ### Docker
 
 ```bash
@@ -80,7 +86,7 @@ want unreleased changes from `main`.
 
 ```bash
 pf-agent quickstart            # 60-second offline tour, no API keys needed
-pf-agent setup                 # guided wizard: providers, keys, workspace
+pf-agent setup --quick         # guided wizard: provider, API key, model, language, workspace
 pf-agent                       # full-screen terminal UI
 pf-agent tui                   # same TUI, with explicit flags
 pf-agent chat --message "draft an opening" --provider fake
@@ -90,13 +96,13 @@ pf-agent service serve         # local HTTP API on 127.0.0.1:8765
 
 The bare `pf-agent` command works like other coding agents' CLIs: the
 first run bootstraps a workspace, then opens the full-screen terminal UI.
-After `pf-agent setup` it talks to your configured default model. Without
+After `pf-agent setup --quick` it talks to your configured default model. Without
 setup, it runs against the offline fake provider so you can tour the CLI
 with no keys.
 
-Set provider API keys in your environment (or `.env`, see
-`.env.example`): `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, and friends.
-Every command accepts `--provider fake` for a no-key offline tour.
+The setup wizard writes API keys to `.pf-agent/.env`; `config.yaml` stores
+only provider metadata and `api_key_ref`. Every command accepts
+`--provider fake` for a no-key offline tour.
 
 ## Project Boundary
 
@@ -108,9 +114,9 @@ workflow state, conversational agent loops, and background event handling.
 
 ## Release Status
 
-`v0.3.1` was published on July 3, 2026 through PyPI Trusted Publishing.
-Release links: [PyPI](https://pypi.org/project/proseforge-agent/0.3.1/),
-[GitHub Release](https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.1),
+`v0.3.2` is prepared as the first-run UX patch release for July 3, 2026.
+Release links: [PyPI](https://pypi.org/project/proseforge-agent/0.3.2/),
+[GitHub Release](https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.2),
 [GHCR image](https://github.com/remacheybn408-boop/ProseForge-Agent/pkgs/container/proseforge-agent),
 and [`CHANGELOG.md`](CHANGELOG.md).
 
@@ -304,7 +310,7 @@ requests re-checked by downstream policy, plus research-ready
 `PyPIPublisher` — real `python -m build` + `twine upload` for TestPyPI /
 PyPI, gated by the Task 47 `PackageChecker` and duplicate-version refusal,
 with injectable runner + `dry_run` for offline tests; tokens never appear
-in the plan, report, or logs. The public `v0.3.1` release path uses PyPI
+in the plan, report, or logs. The public `v0.3.2` release path uses PyPI
 Trusted Publishing / OIDC, so no long-lived PyPI token is stored in the
 repository.
 

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "proseforge-agent"
-version = "0.3.1"
+version = "0.3.2"
 description = "Agentic orchestration layer for long-form novel production on top of the ProseForge engine."
 readme = "README.md"
 requires-python = ">=3.10"

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/src/proseforge_agent/__init__.py

@@ -5,6 +5,6 @@ or require API keys. Subsystem behavior lives behind dedicated submodules
 that are imported explicitly by their callers.
 """
 
-__version__ = "0.3.1"
+__version__ = "0.3.2"
 
 __all__ = ["__version__"]

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/src/proseforge_agent/cli.py

@@ -5536,6 +5536,8 @@ def _handle_setup(args: argparse.Namespace) -> int:
         no_shell=args.no_shell,
         repair=args.repair,
         print_config=args.print_config,
+        interactive=not args.non_interactive and not args.minimal and not args.print_config,
+        launch=False,
     )
     if args.print_config:
         print(result.rendered_config, end="")

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/src/proseforge_agent/dotenv.py

@@ -7,7 +7,8 @@ shell variables. Load precedence (higher wins):
 1. existing ``os.environ`` (never overridden by default)
 2. ``.env.local``
 3. ``.env``
-4. ``<config_dir>/proseforge-agent.env`` (machine defaults)
+4. ``.pf-agent/.env`` (workspace setup)
+5. ``<config_dir>/proseforge-agent.env`` (machine defaults)
 
 ``DotenvLoader`` takes an ordered list of paths from LOWEST to HIGHEST
 precedence and merges them; ``load_default_files`` builds that list.
@@ -99,10 +100,11 @@ class DotenvLoader:
 
 def default_files(root: str | Path = ".", config_dir: str | Path | None = None) -> list[Path]:
     """Ordered low-to-high precedence file list for the default load."""
-    root_path = Path(root)
+    root_path = Path(root).resolve()
     files: list[Path] = []
     if config_dir is not None:
         files.append(Path(config_dir) / _MACHINE_ENV_NAME)
+    files.append(root_path / ".pf-agent" / ".env")
     files.append(root_path / ".env")
     files.append(root_path / ".env.local")
     return files

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/src/proseforge_agent/setup/wizard.py

@@ -3,9 +3,10 @@
 from __future__ import annotations
 
 import os
+import sys
 from dataclasses import asdict, dataclass, field
 from pathlib import Path
-from typing import Any, Literal
+from typing import Any, Literal, TextIO
 
 from .config_generator import SetupConfigGenerator, load_existing, provider_entry, redact_config_text
 from .modes import PROVIDER_ORDER, SetupMode
@@ -58,11 +59,15 @@ class SetupWizard:
         root: str | Path | None = None,
         *,
         env: dict[str, str] | None = None,
+        input_stream: TextIO | None = None,
+        output_stream: TextIO | None = None,
         provider_ping=None,
         config_generator: SetupConfigGenerator | None = None,
     ) -> None:
         self.root = Path(root or ".pf-agent")
         self.env = dict(os.environ if env is None else env)
+        self.input_stream = input_stream or sys.stdin
+        self.output_stream = output_stream or sys.stdout
         self.provider_ping = provider_ping
         self.config_generator = config_generator or SetupConfigGenerator()
 
@@ -76,10 +81,14 @@ class SetupWizard:
         no_shell: bool = False,
         repair: bool = False,
         print_config: bool = False,
+        interactive: bool = False,
+        launch: bool = True,
     ) -> SetupResult:
         mode = SetupMode(mode)
         config_path = self.root / "config.yaml"
         workspace_path = self.root / "workspace"
+        provider_models: dict[str, str] = {}
+        env_values: dict[str, str] = {}
 
         if print_config and config_path.exists():
             text = redact_config_text(config_path.read_text(encoding="utf-8"))
@@ -97,9 +106,26 @@ class SetupWizard:
         needs_backup = bool(config_path.exists() and (reconfigure or add_provider or repair))
         backup_path = backup_config(config_path) if needs_backup else None
         existing = load_existing(config_path)
+
+        interactive_provider: str | None = None
+        if interactive and mode != SetupMode.MINIMAL and not add_provider:
+            payload = self._collect_interactive_setup(default_workspace=workspace_path)
+            interactive_provider = payload["provider"]
+            workspace_path = payload["workspace_path"]
+            provider_models.update(payload["provider_models"])
+            env_values.update(payload["env_values"])
+            self.env.update({key: value for key, value in env_values.items() if value})
+            existing = {
+                **existing,
+                "agent": {
+                    **dict(existing.get("agent") or {}),
+                    **payload["agent"],
+                },
+            }
+
         ensure_workspace(workspace_path)
 
-        selected = self._select_providers(mode, add_provider=add_provider)
+        selected = self._select_providers(mode, add_provider=add_provider, interactive_provider=interactive_provider)
         providers: list[ProviderSetupResult] = []
         provider_config: dict[str, dict[str, Any]] = {}
         for name in selected:
@@ -109,6 +135,7 @@ class SetupWizard:
                 name,
                 configured=result.configured,
                 key_ref=_key_ref_for(name),
+                model=provider_models.get(name),
                 status=result.status,
             )
             if result.status in {"warn", "fail"} and result.reason:
@@ -137,10 +164,13 @@ class SetupWizard:
         self.root.mkdir(parents=True, exist_ok=True)
         config_text = self.config_generator.render(config)
         config_path.write_text(config_text, encoding="utf-8")
-        (self.root / ".env").write_text(
-            self.config_generator.render_env_template(provider_config),
-            encoding="utf-8",
-        )
+        (self.root / ".env").write_text(self._render_env_file(provider_config, env_values), encoding="utf-8")
+
+        next_steps = ["pf-agent doctor"]
+        if interactive and launch:
+            next_steps.append("pf-agent tui")
+        else:
+            next_steps.append('pf-agent chat --provider fake --message "hello"')
 
         return SetupResult(
             completed=not errors,
@@ -150,17 +180,23 @@ class SetupWizard:
             providers=providers,
             warnings=warnings,
             errors=errors,
-            next_steps=[
-                "pf-agent doctor",
-                "pf-agent chat --provider fake --message \"hello\"",
-            ],
+            next_steps=next_steps,
             backup_path=backup_path,
             rendered_config=redact_config_text(config_text),
         )
 
-    def _select_providers(self, mode: SetupMode, *, add_provider: str | None) -> list[str]:
+    def _select_providers(
+        self,
+        mode: SetupMode,
+        *,
+        add_provider: str | None,
+        interactive_provider: str | None = None,
+    ) -> list[str]:
         if add_provider:
             return [_normalize_provider(add_provider), "fake"]
+        if interactive_provider:
+            provider = _normalize_provider(interactive_provider)
+            return ["fake"] if provider == "fake" else [provider, "fake"]
         if mode == SetupMode.MINIMAL:
             return ["fake"]
         if self.env.get("DEEPSEEK_API_KEY"):
@@ -189,6 +225,74 @@ class SetupWizard:
             return ProviderSetupResult(name, True, configured, "warn", reason=str(exc))
         return ProviderSetupResult(name, True, configured, "ok", latency_ms=int(latency or 0))
 
+    def _collect_interactive_setup(self, *, default_workspace: Path) -> dict[str, Any]:
+        self._say("ProseForge Agent setup")
+        self._say("Choose provider:")
+        choices = [
+            ("openai", "OpenAI"),
+            ("anthropic", "Anthropic"),
+            ("deepseek", "DeepSeek"),
+            ("qwen", "Qwen"),
+            ("gemini", "Gemini"),
+            ("fake", "Fake offline"),
+            ("other", "Other bundled provider"),
+        ]
+        for index, (_, label) in enumerate(choices, start=1):
+            self._say(f"  {index}. {label}")
+        raw_provider = self._ask("Provider [OpenAI]", default="openai")
+        provider = _provider_choice(raw_provider, choices)
+        if provider == "other":
+            provider = _normalize_provider(self._ask("Provider id", default="deepseek"))
+
+        env_values: dict[str, str] = {}
+        provider_models: dict[str, str] = {}
+        if provider != "fake":
+            env_key = _env_key_for(provider)
+            api_key = self._ask(f"API key ({env_key}, leave blank for offline fake)")
+            env_values[env_key] = api_key
+            model = self._ask("Model", default=_default_model_for(provider))
+            provider_models[provider] = model
+        else:
+            self._say("API key: skipped for fake offline provider")
+
+        language = self._ask("Language", default="zh-CN")
+        persona = self._ask("Persona", default="default")
+        workspace_text = self._ask("Workspace path", default=str(default_workspace))
+        self._ask("ProseForge root (optional)")
+
+        return {
+            "provider": provider,
+            "workspace_path": Path(workspace_text or default_workspace),
+            "provider_models": provider_models,
+            "env_values": env_values,
+            "agent": {"language": language or "zh-CN", "profile": persona or "default"},
+        }
+
+    def _ask(self, prompt: str, default: str = "") -> str:
+        suffix = f" [{default}]" if default else ""
+        self.output_stream.write(f"{prompt}{suffix}: ")
+        self.output_stream.flush()
+        line = self.input_stream.readline()
+        if line == "":
+            self.output_stream.write("\n")
+            return default
+        value = line.strip()
+        return value or default
+
+    def _say(self, text: str) -> None:
+        self.output_stream.write(text + "\n")
+
+    def _render_env_file(self, providers: dict[str, dict[str, Any]], env_values: dict[str, str]) -> str:
+        lines = ["# ProseForge Agent environment overrides. Do not commit real keys."]
+        keys: list[str] = []
+        for name in sorted(providers):
+            if name == "fake":
+                continue
+            key = _env_key_for(name)
+            keys.append(f"{key}={env_values.get(key, '')}")
+        lines.extend(keys or ["# No external provider keys required."])
+        return "\n".join(lines) + "\n"
+
 
 def _normalize_provider(name: str) -> str:
     return (name or "deepseek").lower().strip()
@@ -204,6 +308,29 @@ def _key_ref_for(provider: str) -> str | None:
     return f"env://{_env_key_for(provider)}"
 
 
+def _provider_choice(value: str, choices: list[tuple[str, str]]) -> str:
+    raw = (value or "openai").strip().lower()
+    if raw.isdigit():
+        index = int(raw) - 1
+        if 0 <= index < len(choices):
+            return choices[index][0]
+    for provider, label in choices:
+        if raw in {provider, label.lower()}:
+            return provider
+    return _normalize_provider(raw)
+
+
+def _default_model_for(provider: str) -> str:
+    defaults = {
+        "openai": "gpt-4o-mini",
+        "anthropic": "claude-3-5-sonnet-latest",
+        "deepseek": "deepseek-chat",
+        "qwen": "qwen-plus",
+        "gemini": "gemini-1.5-flash",
+    }
+    return defaults.get(provider, f"{provider}-chat")
+
+
 __all__ = [
     "PROVIDER_ORDER",
     "ProviderSetupResult",

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/src/proseforge_agent/tui/fullscreen.py

@@ -170,12 +170,39 @@ class FullScreenApp(TerminalApp):
         ]
         return self.style.meta(" · ".join(parts)[:width])
 
+    def _welcome_lines(self, width: int) -> list[str]:
+        if width < 48:
+            raw = [
+                "Ready",
+                "Try: /help",
+                "pf-agent setup",
+                "Ask for a scene, outline, or revision.",
+            ]
+        else:
+            raw = [
+                "Ready to write.",
+                f"Provider: {self.state.provider}    Model: {self.state.model or 'default'}",
+                "Ask for drafting, revision, planning, or project help.",
+                "",
+                "Try:",
+                "  /help              show commands",
+                "  pf-agent setup     configure a real model",
+                "  pf-agent quickstart run the offline demo",
+            ]
+        lines: list[str] = []
+        for line in raw:
+            wrapped = _wrap(line, max(1, width - 2))
+            lines.extend(("  " + item)[:width] for item in wrapped)
+        return lines
+
     def _render_frame(self, width: int, height: int) -> tuple[list[str], int, int]:
         transcript_rows = max(1, height - 4)
         wrapped: list[str] = []
         pending_lines = self._pending.split("\n") if self._pending else []
         for line in [*self.transcript, *pending_lines]:
             wrapped.extend(_wrap(line, width))
+        if not wrapped:
+            wrapped = self._welcome_lines(width)
         visible = wrapped[-transcript_rows:]
         visible += [""] * (transcript_rows - len(visible))
 
@@ -204,7 +231,7 @@ class FullScreenApp(TerminalApp):
 
         kernel = self._ensure_kernel()
         self._ensure_session()
-        self.transcript.append(f"you: {text}")
+        self.transcript.append(f"You: {text}")
         self.state = replace(self.state, chat_history=(*self.state.chat_history, text)[-20:])
         request = AgentTurnRequest(
             session_id=self.state.session_id,
@@ -215,7 +242,7 @@ class FullScreenApp(TerminalApp):
         )
         try:
             if self.state.stream and hasattr(kernel, "run_turn_stream"):
-                self._pending = "agent: "
+                self._pending = "Agent: "
                 self._paint()
                 for chunk in kernel.run_turn_stream(request):
                     self._pending += chunk.text
@@ -223,13 +250,13 @@ class FullScreenApp(TerminalApp):
                 self.transcript.extend(self._pending.split("\n"))
             else:
                 result = kernel.run_turn(request)
-                self.transcript.extend(f"agent: {result.text}".split("\n"))
+                self.transcript.extend(f"Agent: {result.text}".split("\n"))
         except KeyboardInterrupt:
             if self._pending:
                 self.transcript.extend(self._pending.split("\n"))
             self.transcript.append("(turn interrupted)")
         except Exception as exc:  # noqa: BLE001 - a failed turn must not kill the app
-            self.transcript.append(f"error: {exc}")
+            self.transcript.append(f"Error: {exc}")
         finally:
             self._pending = ""
         self._refresh_usage()

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2/src/proseforge_agent.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: proseforge-agent
-Version: 0.3.1
+Version: 0.3.2
 Summary: Agentic orchestration layer for long-form novel production on top of the ProseForge engine.
 Author: ProseForge Agent
 License: Apache-2.0
@@ -32,7 +32,7 @@ Dynamic: license-file
 <p align="center">
   <a href="README.zh-CN.md">简体中文</a> ·
   <a href="https://pypi.org/project/proseforge-agent/">PyPI</a> ·
-  <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.1">Release v0.3.1</a> ·
+  <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.2">Release v0.3.2</a> ·
   <a href="https://github.com/remacheybn408-boop/ProseForge-Agent/pkgs/container/proseforge-agent">GHCR</a>
 </p>
 
@@ -59,7 +59,7 @@ image, or a background agent with gateway integrations.
 <tr><td><b>Review and revise</b></td><td>Run a chapter lifecycle from draft to review, rewrite, and accept, with quality gates and rollback-friendly artifacts.</td></tr>
 <tr><td><b>Use the model you want</b></td><td>Profiles for OpenAI, Anthropic, Gemini, xAI/Grok, DeepSeek, Qwen, GLM, Mimo, MiniMax, Doubao, plus an offline fake provider.</td></tr>
 <tr><td><b>Run it like a real tool</b></td><td>Full-screen TUI by default, explicit chat command, local API, Docker image, Telegram gateway, MCP stdio, local execution, cron, and telemetry.</td></tr>
-<tr><td><b>Published now</b></td><td><code>v0.3.1</code> is live on PyPI, GHCR, and GitHub Releases. Release gate: <code>1358 tests passing</code> — 35 integration (real subprocess / HTTP / SQLite), the rest offline fakes.</td></tr>
+<tr><td><b>Published now</b></td><td><code>v0.3.2</code> is the first-run UX release: one-line install upgrades the PyPI tool, repairs PATH priority, runs doctor, and opens the guided setup/TUI flow.</td></tr>
 </table>
 
 ---
@@ -85,6 +85,12 @@ curl -fsSL https://raw.githubusercontent.com/remacheybn408-boop/ProseForge-Agent
 iex (irm https://raw.githubusercontent.com/remacheybn408-boop/ProseForge-Agent/main/scripts/install.ps1)
 ```
 
+The one-line installer uses PyPI + `uv tool` by default. It installs or
+upgrades `proseforge-agent`, puts the manager bin directory first on PATH,
+runs `pf-agent doctor`, then starts `pf-agent setup --quick`. Use
+`--no-setup`, `--no-launch`, or `--non-interactive` on POSIX, and
+`-NoSetup`, `-NoLaunch`, or `-NonInteractive` on PowerShell, when scripting.
+
 ### Docker
 
 ```bash
@@ -99,7 +105,7 @@ want unreleased changes from `main`.
 
 ```bash
 pf-agent quickstart            # 60-second offline tour, no API keys needed
-pf-agent setup                 # guided wizard: providers, keys, workspace
+pf-agent setup --quick         # guided wizard: provider, API key, model, language, workspace
 pf-agent                       # full-screen terminal UI
 pf-agent tui                   # same TUI, with explicit flags
 pf-agent chat --message "draft an opening" --provider fake
@@ -109,13 +115,13 @@ pf-agent service serve         # local HTTP API on 127.0.0.1:8765
 
 The bare `pf-agent` command works like other coding agents' CLIs: the
 first run bootstraps a workspace, then opens the full-screen terminal UI.
-After `pf-agent setup` it talks to your configured default model. Without
+After `pf-agent setup --quick` it talks to your configured default model. Without
 setup, it runs against the offline fake provider so you can tour the CLI
 with no keys.
 
-Set provider API keys in your environment (or `.env`, see
-`.env.example`): `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, and friends.
-Every command accepts `--provider fake` for a no-key offline tour.
+The setup wizard writes API keys to `.pf-agent/.env`; `config.yaml` stores
+only provider metadata and `api_key_ref`. Every command accepts
+`--provider fake` for a no-key offline tour.
 
 ## Project Boundary
 
@@ -127,9 +133,9 @@ workflow state, conversational agent loops, and background event handling.
 
 ## Release Status
 
-`v0.3.1` was published on July 3, 2026 through PyPI Trusted Publishing.
-Release links: [PyPI](https://pypi.org/project/proseforge-agent/0.3.1/),
-[GitHub Release](https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.1),
+`v0.3.2` is prepared as the first-run UX patch release for July 3, 2026.
+Release links: [PyPI](https://pypi.org/project/proseforge-agent/0.3.2/),
+[GitHub Release](https://github.com/remacheybn408-boop/ProseForge-Agent/releases/tag/v0.3.2),
 [GHCR image](https://github.com/remacheybn408-boop/ProseForge-Agent/pkgs/container/proseforge-agent),
 and [`CHANGELOG.md`](CHANGELOG.md).
 
@@ -323,7 +329,7 @@ requests re-checked by downstream policy, plus research-ready
 `PyPIPublisher` — real `python -m build` + `twine upload` for TestPyPI /
 PyPI, gated by the Task 47 `PackageChecker` and duplicate-version refusal,
 with injectable runner + `dry_run` for offline tests; tokens never appear
-in the plan, report, or logs. The public `v0.3.1` release path uses PyPI
+in the plan, report, or logs. The public `v0.3.2` release path uses PyPI
 Trusted Publishing / OIDC, so no long-lived PyPI token is stored in the
 repository.
 

{proseforge_agent-0.3.1 → proseforge_agent-0.3.2}/src/proseforge_agent.egg-info/SOURCES.txt

@@ -324,6 +324,7 @@ tests/test_default_provider_resolution.py
 tests/test_default_tui_on_bare_command.py
 tests/test_docker_distribution.py
 tests/test_dotenv_support.py
+tests/test_dotenv_workspace_env.py
 tests/test_e2e_demo.py
 tests/test_evidence_pack.py
 tests/test_execution_guard_timeout.py
@@ -335,6 +336,7 @@ tests/test_general_tools.py
 tests/test_graceful_degradation.py
 tests/test_injection_guard_cjk_precision.py
 tests/test_installation_doctor.py
+tests/test_installer_first_run_ux.py
 tests/test_intent_router.py
 tests/test_linux_native_support.py
 tests/test_llm_fake_and_routing.py

proseforge_agent-0.3.2/tests/test_dotenv_workspace_env.py

@@ -0,0 +1,24 @@
+"""Workspace .pf-agent/.env loading tests."""
+
+from __future__ import annotations
+
+from proseforge_agent.dotenv import default_files, load_default_files
+
+
+def test_default_dotenv_files_include_workspace_env(tmp_path, monkeypatch):
+    monkeypatch.chdir(tmp_path)
+    paths = default_files()
+
+    assert tmp_path / ".pf-agent" / ".env" in paths
+
+
+def test_load_default_files_reads_workspace_env(tmp_path, monkeypatch):
+    monkeypatch.chdir(tmp_path)
+    monkeypatch.delenv("OPENAI_API_KEY", raising=False)
+    (tmp_path / ".pf-agent").mkdir()
+    (tmp_path / ".pf-agent" / ".env").write_text("OPENAI_API_KEY=sk-from-workspace\n", encoding="utf-8")
+
+    loaded = load_default_files()
+
+    assert loaded["OPENAI_API_KEY"] == str(tmp_path / ".pf-agent" / ".env")
+    assert __import__("os").environ["OPENAI_API_KEY"] == "sk-from-workspace"