Setup & onboarding fixes (doctor Playwright check, idempotent init, required text resume)

CHANGELOG.md

@@ -5,6 +5,15 @@ All notable changes to ApplyPilot will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [Unreleased]
+
+### Changed
+- `applypilot doctor` checks for the Playwright browser; README and the init
+  wizard tell you to run `playwright install chromium`.
+- Re-running `applypilot init` merges into the existing `.env` (preserving keys
+  like `CAPSOLVER_API_KEY`) and prompts before overwriting `profile.json` /
+  `searches.yaml`. A plain-text resume is now required (or an explicit skip).
+
 ## [0.2.0] - 2026-02-17
 
 ### Added

README.md

@@ -29,6 +29,7 @@ Three commands. That's it.
 ```bash
 pip install applypilot
 pip install --no-deps python-jobspy && pip install pydantic tls-client requests markdownify regex
+playwright install chromium   # the headless browser used by enrich, smart-extract, and PDF rendering
 applypilot init          # one-time setup: resume, profile, preferences, API keys
 applypilot doctor        # verify your setup — shows what's installed and what's missing
 applypilot run           # discover > enrich > score > tailor > cover letters
@@ -89,6 +90,7 @@ Each stage is independent. Run them all or pick what you need.
 | Component | Required For | Details |
 |-----------|-------------|---------|
 | Python 3.11+ | Everything | Core runtime |
+| Playwright Chromium | Enrich, smart-extract, PDF | `playwright install chromium` after pip install (not bundled) |
 | Node.js 18+ | Auto-apply | Needed for `npx` to run Playwright MCP server |
 | Gemini API key | Scoring, tailoring, cover letters | Free tier (15 RPM / 1M tokens/day) is enough |
 | Chrome/Chromium | Auto-apply | Auto-detected on most systems |

src/applypilot/cli.py

@@ -412,6 +412,21 @@ def doctor() -> None:
         results.append(("Chrome/Chromium", fail_mark,
                         "Install Chrome or set CHROME_PATH env var (needed for auto-apply)"))
 
+    # Playwright browser (separate from Chrome; pip does not install it).
+    # NOTE: os is imported locally above; avoid Path here -- a NameError would be
+    # swallowed by this except and always report MISSING.
+    try:
+        from playwright.sync_api import sync_playwright
+        with sync_playwright() as p:
+            pw_exe = p.chromium.executable_path
+        if pw_exe and os.path.exists(pw_exe):
+            results.append(("Playwright browser", ok_mark, pw_exe))
+        else:
+            raise FileNotFoundError(pw_exe)
+    except Exception:
+        results.append(("Playwright browser", fail_mark,
+                        "Run: playwright install chromium (needed for enrich/smart-extract/PDF)"))
+
     # Node.js / npx (for Playwright MCP)
     npx_bin = shutil.which("npx")
     if npx_bin:

src/applypilot/wizard/init.py

@@ -31,6 +31,35 @@
 console = Console()
 
 
+def _merge_env(existing_text: str, new_pairs: dict) -> str:
+    """Merge new KEY=VALUE pairs into an existing .env, preserving everything else.
+
+    Re-running init must not silently drop keys the user set elsewhere (a
+    CapSolver key, a manual CHROME_PATH). Unknown keys and comments are kept;
+    keys present in new_pairs are overwritten in place.
+    """
+    lines = existing_text.splitlines()
+    out: list[str] = []
+    seen: set[str] = set()
+    for line in lines:
+        stripped = line.strip()
+        if not stripped or stripped.startswith("#") or "=" not in stripped:
+            out.append(line)
+            continue
+        key = stripped.split("=", 1)[0].strip()
+        if key in new_pairs:
+            out.append(f"{key}={new_pairs[key]}")
+            seen.add(key)
+        else:
+            out.append(line)
+    if not out or out[0].strip() != "# ApplyPilot configuration":
+        out.insert(0, "# ApplyPilot configuration")
+    for key, value in new_pairs.items():
+        if key not in seen:
+            out.append(f"{key}={value}")
+    return "\n".join(out) + "\n"
+
+
 # ---------------------------------------------------------------------------
 # Resume
 # ---------------------------------------------------------------------------
@@ -59,18 +88,28 @@ def _setup_resume() -> None:
             shutil.copy2(src, RESUME_PDF_PATH)
             console.print(f"[green]Copied to {RESUME_PDF_PATH}[/green]")
 
-            # Also ask for a plain-text version for LLM consumption
-            txt_path_str = Prompt.ask(
-                "Plain-text version of your resume (.txt)",
-                default="",
-            )
-            if txt_path_str.strip():
-                txt_src = Path(txt_path_str.strip().strip('"').strip("'")).expanduser().resolve()
+            # A plain-text resume is REQUIRED -- scoring, tailoring and cover
+            # letters read it and crash hours later without it. Loop until we
+            # get a valid file or the user explicitly types 'skip'.
+            while True:
+                txt_path_str = Prompt.ask(
+                    "Plain-text version of your resume (.txt path, or 'skip')"
+                ).strip()
+                if txt_path_str.lower() == "skip":
+                    console.print(
+                        f"[red]Scoring, tailoring, and cover letters will FAIL without "
+                        f"resume.txt. Add it later at {RESUME_PATH}.[/red]"
+                    )
+                    break
+                if not txt_path_str:
+                    console.print("[yellow]Enter a .txt path or type 'skip'.[/yellow]")
+                    continue
+                txt_src = Path(txt_path_str.strip('"').strip("'")).expanduser().resolve()
                 if txt_src.exists():
                     shutil.copy2(txt_src, RESUME_PATH)
                     console.print(f"[green]Copied to {RESUME_PATH}[/green]")
-                else:
-                    console.print("[yellow]File not found, skipping plain-text copy.[/yellow]")
+                    break
+                console.print(f"[red]File not found:[/red] {txt_src}")
         break
 
 
@@ -82,6 +121,12 @@ def _setup_profile() -> dict:
     """Walk through profile questions and return a nested profile dict."""
     console.print(Panel("[bold]Step 2: Profile[/bold]\nTell ApplyPilot about yourself. This powers scoring, tailoring, and auto-fill."))
 
+    if PROFILE_PATH.exists() and not Confirm.ask(
+        f"profile.json already exists at {PROFILE_PATH} — overwrite?", default=False
+    ):
+        console.print("[dim]Keeping existing profile.json[/dim]")
+        return json.loads(PROFILE_PATH.read_text(encoding="utf-8"))
+
     profile: dict = {}
 
     # -- Personal --
@@ -188,6 +233,12 @@ def _setup_searches() -> None:
     """Generate a searches.yaml from user input."""
     console.print(Panel("[bold]Step 3: Job Search Config[/bold]\nDefine what you're looking for."))
 
+    if SEARCH_CONFIG_PATH.exists() and not Confirm.ask(
+        f"searches.yaml already exists at {SEARCH_CONFIG_PATH} — overwrite?", default=False
+    ):
+        console.print("[dim]Keeping existing searches.yaml[/dim]")
+        return
+
     location = Prompt.ask("Target location (e.g. 'Remote', 'Canada', 'New York, NY')", default="Remote")
     distance_str = Prompt.ask("Search radius in miles (0 for remote-only)", default="0")
     try:
@@ -252,26 +303,24 @@ def _setup_ai_features() -> None:
         default="gemini",
     )
 
-    env_lines = ["# ApplyPilot configuration", ""]
+    new_pairs: dict[str, str] = {}
 
     if provider == "gemini":
         api_key = Prompt.ask("Gemini API key (from aistudio.google.com)")
-        model = Prompt.ask("Model", default="gemini-2.0-flash")
-        env_lines.append(f"GEMINI_API_KEY={api_key}")
-        env_lines.append(f"LLM_MODEL={model}")
+        new_pairs["GEMINI_API_KEY"] = api_key
+        new_pairs["LLM_MODEL"] = Prompt.ask("Model", default="gemini-2.0-flash")
     elif provider == "openai":
         api_key = Prompt.ask("OpenAI API key")
-        model = Prompt.ask("Model", default="gpt-4o-mini")
-        env_lines.append(f"OPENAI_API_KEY={api_key}")
-        env_lines.append(f"LLM_MODEL={model}")
+        new_pairs["OPENAI_API_KEY"] = api_key
+        new_pairs["LLM_MODEL"] = Prompt.ask("Model", default="gpt-4o-mini")
     elif provider == "local":
-        url = Prompt.ask("Local LLM endpoint URL", default="http://localhost:8080/v1")
-        model = Prompt.ask("Model name", default="local-model")
-        env_lines.append(f"LLM_URL={url}")
-        env_lines.append(f"LLM_MODEL={model}")
+        new_pairs["LLM_URL"] = Prompt.ask("Local LLM endpoint URL", default="http://localhost:8080/v1")
+        new_pairs["LLM_MODEL"] = Prompt.ask("Model name", default="local-model")
 
-    env_lines.append("")
-    ENV_PATH.write_text("\n".join(env_lines), encoding="utf-8")
+    # Merge into any existing .env so previously saved keys (e.g. a CapSolver
+    # key, or a manually added CHROME_PATH) survive a re-run of init.
+    existing = ENV_PATH.read_text(encoding="utf-8") if ENV_PATH.exists() else ""
+    ENV_PATH.write_text(_merge_env(existing, new_pairs), encoding="utf-8")
     console.print(f"[green]AI configuration saved to {ENV_PATH}[/green]")
 
 
@@ -376,6 +425,22 @@ def run_wizard() -> None:
         else:
             tier_lines.append(f"  [dim]✗ Tier {t} — {label}  ({cmds})[/dim]")
 
+    # Warn if the Playwright browser isn't installed (pip doesn't bundle it).
+    import os as _os
+    try:
+        from playwright.sync_api import sync_playwright
+        with sync_playwright() as _p:
+            _pw_exe = _p.chromium.executable_path
+        _pw_ok = bool(_pw_exe and _os.path.exists(_pw_exe))
+    except Exception:
+        _pw_ok = False
+    if not _pw_ok:
+        console.print(
+            "[yellow]Playwright browser not found.[/yellow] "
+            "Run [bold]playwright install chromium[/bold] "
+            "(needed for enrich, smart-extract, and PDF rendering).\n"
+        )
+
     unlock_hint = ""
     if tier == 1:
         unlock_hint = "\n[dim]To unlock Tier 2: configure an LLM API key (re-run [bold]applypilot init[/bold]).[/dim]"

tests/test_init_env_merge.py

@@ -0,0 +1,30 @@
+"""F16: re-running init must not destroy existing .env keys."""
+from applypilot.wizard.init import _merge_env
+
+
+def test_merge_preserves_unknown_keys():
+    out = _merge_env("CAPSOLVER_API_KEY=abc\nGEMINI_API_KEY=old", {"GEMINI_API_KEY": "new"})
+    assert "CAPSOLVER_API_KEY=abc" in out
+    assert "GEMINI_API_KEY=new" in out
+    assert "GEMINI_API_KEY=old" not in out
+    # Each key appears exactly once.
+    assert out.count("CAPSOLVER_API_KEY=") == 1
+    assert out.count("GEMINI_API_KEY=") == 1
+
+
+def test_merge_appends_new_keys():
+    out = _merge_env("CHROME_PATH=/usr/bin/chrome", {"GEMINI_API_KEY": "k"})
+    assert "CHROME_PATH=/usr/bin/chrome" in out
+    assert "GEMINI_API_KEY=k" in out
+
+
+def test_merge_from_empty():
+    out = _merge_env("", {"GEMINI_API_KEY": "k", "LLM_MODEL": "m"})
+    assert "GEMINI_API_KEY=k" in out
+    assert "LLM_MODEL=m" in out
+
+
+def test_merge_keeps_comments():
+    out = _merge_env("# my notes\nCAPSOLVER_API_KEY=x", {"LLM_MODEL": "m"})
+    assert "# my notes" in out
+    assert "CAPSOLVER_API_KEY=x" in out