feat: add frictionless onboarding with praisonai setup wizard and post-install hooks#1453
Merged
MervinPraison merged 4 commits intomainfrom Apr 18, 2026
Merged
Conversation
- Add interactive setup command with provider selection (OpenAI, Anthropic, Google, Ollama) - Support both interactive and non-interactive modes - Secure .env file creation with 600 permissions - Idempotent operation (safe to re-run) - Create ~/.praisonai directory structure (logs, sessions) - Modify install.sh and install.ps1 for post-install onboarding - Add --no-onboard flag to skip wizard - Add PraisonAIConfigError for missing API keys - Update doctor command to suggest 'praisonai setup' - Include comprehensive unit and integration tests Fixes #1451 Co-authored-by: MervinPraison <MervinPraison@users.noreply.github.com>
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
praisonai-triage-agent[bot] has reached the 50-review limit for trial accounts. To continue receiving code reviews, upgrade your plan.
Owner
|
@coderabbitai review |
Owner
|
/review |
Contributor
✅ Actions performedReview triggered.
|
Contributor
|
Important Review skippedBot user detected. To trigger a single review, invoke the ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
📝 WalkthroughWalkthroughThis PR introduces an interactive setup/onboarding feature for PraisonAI, including a new Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant CLI as praisonai setup (CLI)
participant Handler as SetupHandler
participant FileSystem as Filesystem<br/>(.env, config.yaml)
User->>CLI: praisonai setup --provider openai --api-key XXX
CLI->>Handler: execute(non_interactive=True, provider="openai", api_key="XXX")
Handler->>Handler: Validate provider & API key
Handler->>Handler: Assign default model
Handler->>FileSystem: Write ~/.praisonai/.env
Handler->>FileSystem: Write ~/.praisonai/config.yaml
Handler->>FileSystem: Create logs/, sessions/ dirs
Handler-->>CLI: Return exit code 0
CLI-->>User: Setup complete
sequenceDiagram
participant User
participant CLI as praisonai setup (CLI)
participant Handler as SetupHandler
participant RichPrompt as rich.prompt
participant FileSystem as Filesystem
User->>CLI: praisonai setup (interactive)
CLI->>Handler: execute(non_interactive=False)
Handler->>RichPrompt: Prompt provider selection
User->>RichPrompt: Select provider (e.g., openai)
RichPrompt-->>Handler: Return provider
Handler->>RichPrompt: Prompt API key
User->>RichPrompt: Enter API key
RichPrompt-->>Handler: Return API key
Handler->>RichPrompt: Prompt model selection
User->>RichPrompt: Select model
RichPrompt-->>Handler: Return model
Handler->>RichPrompt: Prompt telemetry consent
User->>RichPrompt: Confirm telemetry
RichPrompt-->>Handler: Return telemetry choice
Handler->>FileSystem: Write .env, config.yaml, optionally agents.yaml
Handler-->>CLI: Return exit code 0
CLI-->>User: Setup wizard complete
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related issues
Possibly related PRs
Suggested labels
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Owner
|
@copilot Do a thorough review of this PR. Read ALL existing reviewer comments above from Qodo, Coderabbit, and Gemini first — incorporate their findings. Review areas:
|
Agent-Logs-Url: https://github.com/MervinPraison/PraisonAI/sessions/8c683e28-66c2-42f7-abc2-08efd37b22c7 Co-authored-by: MervinPraison <454862+MervinPraison@users.noreply.github.com>
There was a problem hiding this comment.
praisonai-triage-agent[bot] has reached the 50-review limit for trial accounts. To continue receiving code reviews, upgrade your plan.
Agent-Logs-Url: https://github.com/MervinPraison/PraisonAI/sessions/8c683e28-66c2-42f7-abc2-08efd37b22c7 Co-authored-by: MervinPraison <454862+MervinPraison@users.noreply.github.com>
There was a problem hiding this comment.
praisonai-triage-agent[bot] has reached the 50-review limit for trial accounts. To continue receiving code reviews, upgrade your plan.
Contributor
Completed a full PR review and applied targeted fixes in Key findings:
CI note: I checked workflow runs/logs via Actions API; the failing runs for this branch showed |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 13
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/praisonai-agents/praisonaiagents/errors.py (1)
363-376:⚠️ Potential issue | 🟡 MinorExport
PraisonAIConfigErrorfrompraisonaiagents.errors.The new public error is missing from
__all__, sofrom praisonaiagents.errors import *will omit it despite this list being the file’s “Export all error types” surface.Proposed fix
"HandoffError", "HandoffCycleError", "HandoffDepthError", - "HandoffTimeoutError" + "HandoffTimeoutError", + "PraisonAIConfigError" ]🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/praisonai-agents/praisonaiagents/errors.py` around lines 363 - 376, The __all__ export list in praisonaiagents.errors is missing the new public error class PraisonAIConfigError, so wildcard imports (from praisonaiagents.errors import *) will omit it; update the __all__ list to include "PraisonAIConfigError" alongside the existing symbols (e.g., "PraisonAIError", "ToolExecutionError", "LLMError", etc.) so the new class is exported publicly.
🧹 Nitpick comments (1)
src/praisonai/scripts/install.sh (1)
632-644: Python resolution can produce an empty$pywhenPYTHON_CMDis unset and venv is skipped.When
SKIP_VENV=1(orvenv_diris empty) andPYTHON_CMDis empty (the common case —find_pythonwrites to stdout but its result is never assigned toPYTHON_CMD),pybecomes"", then[[ ! -x "" ]]is true and it falls back topython3. That's fine, but it's fragile: if a user passed--python /custom/pythonearlier,PYTHON_CMDwill be set to that value and used here directly, bypassing the venv's interpreter even when one exists.Suggest flipping the precedence to prefer the venv interpreter first, then
PYTHON_CMD, thenpython3:Proposed fix
- # Determine python command - local py="$PYTHON_CMD" - if [[ -n "$venv_dir" ]] && [[ "$SKIP_VENV" != "1" ]]; then - py="$venv_dir/bin/python" - fi - - # Fallback to system python if venv python doesn't exist - if [[ ! -x "$py" ]]; then - py="python3" - fi + # Prefer venv python, then user-specified, then system python3 + local py="" + if [[ -n "$venv_dir" && "$SKIP_VENV" != "1" && -x "$venv_dir/bin/python" ]]; then + py="$venv_dir/bin/python" + elif [[ -n "$PYTHON_CMD" ]] && command -v "$PYTHON_CMD" >/dev/null 2>&1; then + py="$PYTHON_CMD" + else + py="python3" + fi🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/praisonai/scripts/install.sh` around lines 632 - 644, The current python selection can end up using PYTHON_CMD even when a valid venv interpreter exists or produce an empty $py; change the selection in the block that sets local py so it prefers the venv interpreter first (when venv_dir is set and SKIP_VENV != "1"), then uses PYTHON_CMD only if non-empty and executable, and finally falls back to "python3"; implement this by setting py to the venv path when applicable, otherwise checking [[ -n "$PYTHON_CMD" && -x "$PYTHON_CMD" ]] to assign py="$PYTHON_CMD", and only if neither is available set py="python3" (use the existing variables py, venv_dir, SKIP_VENV, and PYTHON_CMD in the refactor).
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@src/praisonai-agents/praisonaiagents/errors.py`:
- Around line 295-323: PraisonAIConfigError should carry a structured
remediation hint: add an optional parameter remediation_hint: Optional[str] =
None to the PraisonAIConfigError __init__, compute a sensible default (e.g.,
"Verify and set the {config_key} configuration" when config_key is present) when
remediation_hint is not provided, include this hint in the error's context
(context["remediation_hint"] = remediation_hint) and set self.remediation_hint =
remediation_hint, and ensure the call to super().__init__ still receives the
augmented context so downstream renderers and on_error hooks (error interception
points) can show consistent recovery steps; update references to
PraisonAIConfigError accordingly.
In `@src/praisonai/praisonai/cli/commands/setup.py`:
- Around line 152-164: The reset currently only unlinks ENV_FILE (deletes .env)
but must also remove the written config.yaml; update the reset flow in setup.py
so after checking/removing ENV_FILE it also checks for the config file (e.g.,
the CONFIG_FILE or CONFIG_YAML path used elsewhere in this module, or fallback
to "config.yaml") and unlinks it if it exists, handling errors the same way and
updating output.print_success to reflect both files removed (and still
respecting the force/confirm logic and try/except around file deletion).
In `@src/praisonai/praisonai/cli/features/setup/handler.py`:
- Around line 227-228: The starter YAML currently uses a hard-coded model in
AGENTS_TEMPLATE (gpt-4o-mini) causing mismatches; update the workflow so the
chosen model is threaded into YAML generation: change the caller that sets
create_starter (where create_starter is used) to pass the selected model to
_create_starter_yaml, update _create_starter_yaml to accept a model parameter
and substitute it into AGENTS_TEMPLATE rather than using the hard-coded value,
and ensure other code paths noted around the block at lines ~299-307 also pass
the selected model when generating the YAML so CLI, Python API, and YAML outputs
stay consistent.
- Around line 64-69: The directory creation in _create_directories currently
uses mkdir() without explicit permissions; change it to create ~/.praisonai,
logs, and sessions with owner-only perms by passing mode=0o700 to mkdir (for
praison_home, praison_home/"logs", and praison_home/"sessions") and then ensure
existing directories also get tightened by calling praison_home.chmod(0o700) and
likewise chmod(0o700) on the logs and sessions paths after creation; keep the
output.print_success line unchanged so the success message still prints.
- Line 127: Ruff flags F541 because there are several output.console.print calls
using unnecessary f-strings; locate each output.console.print(...) invocation
(e.g., the line with output.console.print(f'Try: [cyan]praisonai "Say hello in
one sentence"[/cyan]') and the other occurrences around the reported ranges) and
remove the leading f from the string literal so they become plain strings (e.g.,
change f'...' to '...') since no interpolation is performed; ensure you update
all occurrences referenced (lines ~127, 192, 202, 211, 231-238, 304-309) so no
placeholder-free f-strings remain.
- Around line 102-120: The current non-interactive setup allows passing secrets
via --api-key and can write model: null for custom/no-default providers; change
the logic in the handler to (1) prefer reading the secret from the environment
using env_key (os.getenv(env_key)) and only use the api_key CLI value as a
fallback with an explicit warning or reject it in non-interactive runs to avoid
exposing secrets in process args, (2) when env_key exists and env value is
present, populate env_vars[env_key] from that value (do not write raw
--api-key), (3) enforce that if default_model is None or provider == "custom"
and model is not provided then return an error asking for --model (do not set
model to None in config), and (4) ensure the final config dict uses the resolved
non-null model and includes provider, model, telemetry, and env_vars
appropriately (references: env_key, api_key, provider, provider_name, model,
default_model, env_vars, config).
- Around line 269-271: The current sequence calls
env_file.write_text(env_content) then env_file.chmod(0o600), which briefly
creates the .env with umask-controlled permissions; instead open and create the
file with the desired mode atomically so it is never writable/readable with
permissive bits: use os.open on the env_file path with flags
O_WRONLY|O_CREAT|O_TRUNC and mode 0o600, write env_content (as bytes) via the
returned file descriptor and close it (or wrap the fd in os.fdopen) rather than
calling write_text then chmod; update the code in handler.py where env_file and
env_content are used (replace write_text/chmod calls) to perform this
secure-create-and-write sequence.
In `@src/praisonai/praisonai/cli/features/setup/templates.py`:
- Around line 39-62: The AGENTS_TEMPLATE currently hardcodes llm: "gpt-4o-mini",
which breaks non-OpenAI providers; update AGENTS_TEMPLATE to accept a model
placeholder (e.g., "{model}" or leave llm empty) and change the
_create_starter_yaml function to substitute the selected provider/model from the
setup wizard into that placeholder (or omit the llm key so default
provider/model resolution applies) so the generated agents.yaml uses the
user-selected model instead of always gpt-4o-mini.
In `@src/praisonai/scripts/install.ps1`:
- Around line 426-453: Invoke-Onboarding currently only checks -NoOnboard and
-DryRun and can still run in non-interactive contexts; update Invoke-Onboarding
to also honor an explicit -NoPrompt switch and the environment variable
$env:PRAISONAI_NO_PROMPT, and add a TTY/user-interactive guard using
[Environment]::UserInteractive and [Console]::IsInputRedirected (or similar) so
the function returns early when running in CI, scheduled tasks, or piped install
scenarios; keep references to the existing Invoke-Onboarding function and the
NoOnboard/DryRun checks and ensure the new checks run before starting the
interactive setup wizard.
- Around line 455-468: The try/catch around running the setup wizard won't catch
non-terminating failures from the native python process; after invoking &
$pythonCmd -m praisonai setup, check $LASTEXITCODE and treat any non-zero value
as failure: move the success messages behind a conditional that verifies
$LASTEXITCODE -eq 0 and run the existing warning/help text when it's non-zero
(the same handling currently in the catch block); keep the catch for true
terminating errors (e.g., missing python) but ensure $pythonCmd and the
praisonai setup invocation are followed by an explicit $LASTEXITCODE check to
decide success vs failure.
In `@src/praisonai/scripts/install.sh`:
- Around line 644-655: The current invocation in install.sh that runs the setup
wizard (the if "$py" -m praisonai setup < /dev/tty; then ... in run_onboarding)
only re-attaches stdin to /dev/tty and can still leave stdout/stderr non-TTY
(causing Rich prompts to fall back and a misleading success message); update
run_onboarding to either (A) also attach stdout and stderr to the TTY when
launching the wizard (redirect >/dev/tty and 2>/dev/tty along with the existing
stdin redirection) or (B) perform an explicit TTY check before launching (e.g.
verify [ -t 0 ] && [ -t 1 ] && [ -t 2 ] or at minimum [ -t 1 ]) and skip or warn
if not all required streams are TTYs so the wizard is not invoked in a
broken/non-interactive environment.
In `@src/praisonai/tests/integration/test_setup_integration.py`:
- Around line 8-11: Tests currently use a hard-coded GitHub runner path and a
plain "python" invocation; change subprocess calls to use sys.executable and
compute the working directory relative to the test file using
Path(__file__).resolve() (e.g., package_dir =
Path(__file__).resolve().parents[1] or appropriate parent to reach the
src/praisonai package) and pass cwd=str(package_dir) to subprocess.run/ Popen;
update all occurrences referenced in the file (lines ~33-48, 65-73, 83-97,
121-149) and replace "python" with sys.executable to ensure the current test
interpreter is used, and where external state, timing or network calls are
involved wrap/replace them with mocks/fixtures (use unittest.mock or pytest
fixtures) so the subprocess tests are deterministic and runner-path independent.
In `@src/praisonai/tests/unit/cli/test_setup_command.py`:
- Around line 69-70: The permission assertion using
oct(env_file.stat().st_mode)[-3:] == "600" will fail on Windows; update the
test(s) that check file permissions (the assertion around
env_file.stat().st_mode and the similar block at lines 216-230) to skip on
Windows using pytest.mark.skipif(os.name == "nt" or
sys.platform.startswith("win"), reason="permission bits not reliable on
Windows") or alternatively replace the exact-mode assertion with a
platform-aware check (e.g., only assert presence of owner read/write on POSIX
via os.name == "posix" guard before checking st_mode). Ensure you reference the
existing assertion expression env_file.stat().st_mode when locating and
modifying the tests.
---
Outside diff comments:
In `@src/praisonai-agents/praisonaiagents/errors.py`:
- Around line 363-376: The __all__ export list in praisonaiagents.errors is
missing the new public error class PraisonAIConfigError, so wildcard imports
(from praisonaiagents.errors import *) will omit it; update the __all__ list to
include "PraisonAIConfigError" alongside the existing symbols (e.g.,
"PraisonAIError", "ToolExecutionError", "LLMError", etc.) so the new class is
exported publicly.
---
Nitpick comments:
In `@src/praisonai/scripts/install.sh`:
- Around line 632-644: The current python selection can end up using PYTHON_CMD
even when a valid venv interpreter exists or produce an empty $py; change the
selection in the block that sets local py so it prefers the venv interpreter
first (when venv_dir is set and SKIP_VENV != "1"), then uses PYTHON_CMD only if
non-empty and executable, and finally falls back to "python3"; implement this by
setting py to the venv path when applicable, otherwise checking [[ -n
"$PYTHON_CMD" && -x "$PYTHON_CMD" ]] to assign py="$PYTHON_CMD", and only if
neither is available set py="python3" (use the existing variables py, venv_dir,
SKIP_VENV, and PYTHON_CMD in the refactor).
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 5ba87177-287a-48d2-83f2-519bb79c9f86
📒 Files selected for processing (13)
src/praisonai-agents/praisonaiagents/__init__.pysrc/praisonai-agents/praisonaiagents/errors.pysrc/praisonai/praisonai/cli/app.pysrc/praisonai/praisonai/cli/commands/__init__.pysrc/praisonai/praisonai/cli/commands/setup.pysrc/praisonai/praisonai/cli/features/doctor/checks/env_checks.pysrc/praisonai/praisonai/cli/features/setup/__init__.pysrc/praisonai/praisonai/cli/features/setup/handler.pysrc/praisonai/praisonai/cli/features/setup/templates.pysrc/praisonai/scripts/install.ps1src/praisonai/scripts/install.shsrc/praisonai/tests/integration/test_setup_integration.pysrc/praisonai/tests/unit/cli/test_setup_command.py
Comment on lines
+295
to
+323
| class PraisonAIConfigError(PraisonAIError): | ||
| """ | ||
| Configuration error (missing API keys, invalid settings, etc). | ||
|
|
||
| Usually indicates setup issues that require user action. | ||
| """ | ||
|
|
||
| def __init__( | ||
| self, | ||
| message: str, | ||
| config_key: Optional[str] = None, | ||
| agent_id: str = "unknown", | ||
| run_id: Optional[str] = None, | ||
| is_retryable: bool = False, # Config errors need user intervention | ||
| context: Optional[Dict[str, Any]] = None | ||
| ): | ||
| context = context or {} | ||
| if config_key: | ||
| context.update({"config_key": config_key}) | ||
|
|
||
| super().__init__( | ||
| message, | ||
| agent_id=agent_id, | ||
| run_id=run_id, | ||
| error_category="validation", # Configuration is a type of validation error | ||
| is_retryable=is_retryable, | ||
| context=context | ||
| ) | ||
| self.config_key = config_key |
Contributor
There was a problem hiding this comment.
Add a structured remediation hint to config errors.
PraisonAIConfigError captures config_key, but the rendered exception still relies entirely on caller-supplied text for next steps. Add an optional/default remediation hint so setup failures consistently tell users how to recover. As per coding guidelines, "Error handling: Fail fast with clear error messages; include remediation hints in exceptions; propagate context (agent name, tool name, session ID); provide hook points for error interception (on_error events)".
Proposed refinement
class PraisonAIConfigError(PraisonAIError):
@@
def __init__(
self,
message: str,
config_key: Optional[str] = None,
agent_id: str = "unknown",
run_id: Optional[str] = None,
is_retryable: bool = False, # Config errors need user intervention
+ remediation_hint: Optional[str] = None,
context: Optional[Dict[str, Any]] = None
):
context = context or {}
if config_key:
context.update({"config_key": config_key})
+ if remediation_hint is None:
+ remediation_hint = f"Set {config_key} or run the setup wizard before retrying."
+ if remediation_hint:
+ context["remediation_hint"] = remediation_hint
+ message = f"{message} Remediation: {remediation_hint}"
super().__init__(
message,🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai-agents/praisonaiagents/errors.py` around lines 295 - 323,
PraisonAIConfigError should carry a structured remediation hint: add an optional
parameter remediation_hint: Optional[str] = None to the PraisonAIConfigError
__init__, compute a sensible default (e.g., "Verify and set the {config_key}
configuration" when config_key is present) when remediation_hint is not
provided, include this hint in the error's context (context["remediation_hint"]
= remediation_hint) and set self.remediation_hint = remediation_hint, and ensure
the call to super().__init__ still receives the augmented context so downstream
renderers and on_error hooks (error interception points) can show consistent
recovery steps; update references to PraisonAIConfigError accordingly.
Comment on lines
+152
to
+164
| if not ENV_FILE.exists(): | ||
| output.print_info("No setup configuration to reset.") | ||
| return | ||
|
|
||
| if not force: | ||
| confirm = typer.confirm(f"Reset configuration at {ENV_FILE}?") | ||
| if not confirm: | ||
| output.print_info("Reset cancelled.") | ||
| return | ||
|
|
||
| try: | ||
| ENV_FILE.unlink() | ||
| output.print_success("Configuration reset successfully.") |
Contributor
There was a problem hiding this comment.
Reset should remove config.yaml as well as .env.
Setup writes both .env and config.yaml, but reset only deletes .env. A user can reset successfully and still retain stale provider/model/telemetry configuration.
Proposed fix
- if not ENV_FILE.exists():
+ praison_home = get_praison_home()
+ env_file = praison_home / ".env"
+ config_file = praison_home / "config.yaml"
+ files_to_remove = [path for path in (env_file, config_file) if path.exists()]
+
+ if not files_to_remove:
output.print_info("No setup configuration to reset.")
return
if not force:
- confirm = typer.confirm(f"Reset configuration at {ENV_FILE}?")
+ confirm = typer.confirm(f"Reset configuration at {praison_home}?")
if not confirm:
output.print_info("Reset cancelled.")
return
try:
- ENV_FILE.unlink()
+ for path in files_to_remove:
+ path.unlink()
output.print_success("Configuration reset successfully.")🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/praisonai/cli/commands/setup.py` around lines 152 - 164, The
reset currently only unlinks ENV_FILE (deletes .env) but must also remove the
written config.yaml; update the reset flow in setup.py so after
checking/removing ENV_FILE it also checks for the config file (e.g., the
CONFIG_FILE or CONFIG_YAML path used elsewhere in this module, or fallback to
"config.yaml") and unlinks it if it exists, handling errors the same way and
updating output.print_success to reflect both files removed (and still
respecting the force/confirm logic and try/except around file deletion).
Comment on lines
+64
to
+69
| def _create_directories(self, praison_home: Path, output) -> None: | ||
| """Create the ~/.praisonai directory structure.""" | ||
| praison_home.mkdir(parents=True, exist_ok=True) | ||
| (praison_home / "logs").mkdir(exist_ok=True) | ||
| (praison_home / "sessions").mkdir(exist_ok=True) | ||
| output.print_success(f"Created directory structure at {praison_home}") |
Contributor
There was a problem hiding this comment.
Create PraisonAI state directories with private permissions.
mkdir() currently relies on the process umask, so ~/.praisonai, logs, and sessions may become 755. These directories can hold user prompts, outputs, and session metadata; default them to owner-only access.
Proposed fix
def _create_directories(self, praison_home: Path, output) -> None:
"""Create the ~/.praisonai directory structure."""
- praison_home.mkdir(parents=True, exist_ok=True)
- (praison_home / "logs").mkdir(exist_ok=True)
- (praison_home / "sessions").mkdir(exist_ok=True)
+ praison_home.mkdir(parents=True, exist_ok=True, mode=0o700)
+ (praison_home / "logs").mkdir(exist_ok=True, mode=0o700)
+ (praison_home / "sessions").mkdir(exist_ok=True, mode=0o700)
+ praison_home.chmod(0o700)
+ (praison_home / "logs").chmod(0o700)
+ (praison_home / "sessions").chmod(0o700)
output.print_success(f"Created directory structure at {praison_home}")📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| def _create_directories(self, praison_home: Path, output) -> None: | |
| """Create the ~/.praisonai directory structure.""" | |
| praison_home.mkdir(parents=True, exist_ok=True) | |
| (praison_home / "logs").mkdir(exist_ok=True) | |
| (praison_home / "sessions").mkdir(exist_ok=True) | |
| output.print_success(f"Created directory structure at {praison_home}") | |
| def _create_directories(self, praison_home: Path, output) -> None: | |
| """Create the ~/.praisonai directory structure.""" | |
| praison_home.mkdir(parents=True, exist_ok=True, mode=0o700) | |
| (praison_home / "logs").mkdir(exist_ok=True, mode=0o700) | |
| (praison_home / "sessions").mkdir(exist_ok=True, mode=0o700) | |
| praison_home.chmod(0o700) | |
| (praison_home / "logs").chmod(0o700) | |
| (praison_home / "sessions").chmod(0o700) | |
| output.print_success(f"Created directory structure at {praison_home}") |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/praisonai/cli/features/setup/handler.py` around lines 64 - 69,
The directory creation in _create_directories currently uses mkdir() without
explicit permissions; change it to create ~/.praisonai, logs, and sessions with
owner-only perms by passing mode=0o700 to mkdir (for praison_home,
praison_home/"logs", and praison_home/"sessions") and then ensure existing
directories also get tightened by calling praison_home.chmod(0o700) and likewise
chmod(0o700) on the logs and sessions paths after creation; keep the
output.print_success line unchanged so the success message still prints.
Comment on lines
+102
to
+120
| # Validate API key requirement | ||
| if env_key and not api_key: | ||
| output.print_error(f"API key is required for {provider_name}") | ||
| return 1 | ||
|
|
||
| # Use default model if not specified | ||
| if not model: | ||
| model = default_model | ||
|
|
||
| # Write configuration | ||
| env_vars = {} | ||
| if env_key and api_key: | ||
| env_vars[env_key] = api_key | ||
|
|
||
| config = { | ||
| "provider": provider, | ||
| "model": model, | ||
| "telemetry": True, # Default enabled in non-interactive mode | ||
| } |
Contributor
There was a problem hiding this comment.
Tighten non-interactive validation for secrets and custom providers.
CI users currently have to pass API keys via --api-key, which exposes secrets through shell history/process args. Also, custom without --model writes model: null. Read the provider env var as a fallback and require a model when no default exists.
Proposed fix
provider_name = provider_info[0]
env_key = provider_info[1]
default_model = provider_info[2]
# Validate API key requirement
- if env_key and not api_key:
- output.print_error(f"API key is required for {provider_name}")
- return 1
+ if env_key:
+ api_key = api_key or os.getenv(env_key)
+ if not api_key:
+ output.print_error(
+ f"API key is required for {provider_name}; pass --api-key or set {env_key}"
+ )
+ return 1
# Use default model if not specified
if not model:
- model = default_model
+ if default_model:
+ model = default_model
+ else:
+ output.print_error("Model is required for custom provider")
+ return 1🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/praisonai/cli/features/setup/handler.py` around lines 102 -
120, The current non-interactive setup allows passing secrets via --api-key and
can write model: null for custom/no-default providers; change the logic in the
handler to (1) prefer reading the secret from the environment using env_key
(os.getenv(env_key)) and only use the api_key CLI value as a fallback with an
explicit warning or reject it in non-interactive runs to avoid exposing secrets
in process args, (2) when env_key exists and env value is present, populate
env_vars[env_key] from that value (do not write raw --api-key), (3) enforce that
if default_model is None or provider == "custom" and model is not provided then
return an error asking for --model (do not set model to None in config), and (4)
ensure the final config dict uses the resolved non-null model and includes
provider, model, telemetry, and env_vars appropriately (references: env_key,
api_key, provider, provider_name, model, default_model, env_vars, config).
|
|
||
| output.print_success("Setup complete") | ||
| output.console.print(f"Configuration saved to {praison_home}") | ||
| output.console.print(f'Try: [cyan]praisonai "Say hello in one sentence"[/cyan]') |
Contributor
There was a problem hiding this comment.
Remove placeholder-free f-strings flagged by Ruff.
These strings do not interpolate anything and will fail Ruff F541 in lint-enforced CI.
Proposed fix
- output.console.print(f'Try: [cyan]praisonai "Say hello in one sentence"[/cyan]')
+ output.console.print('Try: [cyan]praisonai "Say hello in one sentence"[/cyan]')
@@
- output.console.print(f"\n[bold]3. Choose default model:[/bold]")
+ output.console.print("\n[bold]3. Choose default model:[/bold]")
@@
- output.console.print(f"\n[bold]4. Telemetry settings:[/bold]")
+ output.console.print("\n[bold]4. Telemetry settings:[/bold]")
@@
- output.console.print(f"\n[bold]5. Create starter configuration:[/bold]")
+ output.console.print("\n[bold]5. Create starter configuration:[/bold]")
@@
- output.console.print(f"\n[bold green]✅ Setup complete![/bold green]")
+ output.console.print("\n[bold green]✅ Setup complete![/bold green]")
@@
- output.console.print(f"\n[bold]Next steps:[/bold]")
- output.console.print(f'• Test: [cyan]praisonai "Say hello in one sentence"[/cyan]')
- output.console.print(f'• Chat: [cyan]praisonai chat[/cyan]')
- output.console.print(f'• Help: [cyan]praisonai --help[/cyan]')
- output.console.print(f'• Docs: [cyan]praisonai doctor[/cyan]')
+ output.console.print("\n[bold]Next steps:[/bold]")
+ output.console.print('• Test: [cyan]praisonai "Say hello in one sentence"[/cyan]')
+ output.console.print("• Chat: [cyan]praisonai chat[/cyan]")
+ output.console.print("• Help: [cyan]praisonai --help[/cyan]")
+ output.console.print("• Docs: [cyan]praisonai doctor[/cyan]")
@@
- output.print_warning(f"agents.yaml already exists, skipping creation")
+ output.print_warning("agents.yaml already exists, skipping creation")
@@
- output.console.print(f"Run [cyan]praisonai workflow run --file agents.yaml[/cyan] to test it")
+ output.console.print("Run [cyan]praisonai workflow run --file agents.yaml[/cyan] to test it")Also applies to: 192-192, 202-202, 211-211, 231-238, 304-309
🧰 Tools
🪛 Ruff (0.15.10)
[error] 127-127: f-string without any placeholders
Remove extraneous f prefix
(F541)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/praisonai/cli/features/setup/handler.py` at line 127, Ruff
flags F541 because there are several output.console.print calls using
unnecessary f-strings; locate each output.console.print(...) invocation (e.g.,
the line with output.console.print(f'Try: [cyan]praisonai "Say hello in one
sentence"[/cyan]') and the other occurrences around the reported ranges) and
remove the leading f from the string literal so they become plain strings (e.g.,
change f'...' to '...') since no interpolation is performed; ensure you update
all occurrences referenced (lines ~127, 192, 202, 211, 231-238, 304-309) so no
placeholder-free f-strings remain.
Comment on lines
+426
to
+453
| # Run interactive onboarding after installation | ||
| function Invoke-Onboarding { | ||
| param([string]$VenvPath) | ||
|
|
||
| # Skip onboarding if requested | ||
| if ($NoOnboard) { | ||
| Write-Info "Skipping onboarding (-NoOnboard)" | ||
| return | ||
| } | ||
|
|
||
| # Skip onboarding in dry run mode | ||
| if ($DryRun) { | ||
| Write-Info "Dry run mode — skipping onboarding" | ||
| return | ||
| } | ||
|
|
||
| Write-Step "Starting interactive setup wizard..." | ||
|
|
||
| # Determine python command | ||
| $pythonCmd = "python" | ||
| if ($VenvPath -and -not $NoVenv) { | ||
| $pythonCmd = "$VenvPath\Scripts\python.exe" | ||
| } | ||
|
|
||
| # Fallback to system python if venv python doesn't exist | ||
| if (-not (Test-Path $pythonCmd)) { | ||
| $pythonCmd = "python" | ||
| } |
Contributor
There was a problem hiding this comment.
Missing non-interactive/TTY guard — PowerShell counterpart of the shell script's /dev/tty check.
install.sh correctly skips onboarding when there is no TTY or when NO_PROMPT=1. Invoke-Onboarding skips only on -NoOnboard and -DryRun, so running iwr … | iex inside a CI job, a scheduled task, or a remoting session will still attempt to launch the interactive wizard and then hang or fail. Consider also honoring $env:PRAISONAI_NO_PROMPT/a -NoPrompt switch and checking [Environment]::UserInteractive / [Console]::IsInputRedirected before invoking the wizard.
🧰 Tools
🪛 PSScriptAnalyzer (1.25.0)
[warning] Missing BOM encoding for non-ASCII encoded file 'install.ps1'
(PSUseBOMForUnicodeEncodedFile)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/scripts/install.ps1` around lines 426 - 453, Invoke-Onboarding
currently only checks -NoOnboard and -DryRun and can still run in
non-interactive contexts; update Invoke-Onboarding to also honor an explicit
-NoPrompt switch and the environment variable $env:PRAISONAI_NO_PROMPT, and add
a TTY/user-interactive guard using [Environment]::UserInteractive and
[Console]::IsInputRedirected (or similar) so the function returns early when
running in CI, scheduled tasks, or piped install scenarios; keep references to
the existing Invoke-Onboarding function and the NoOnboard/DryRun checks and
ensure the new checks run before starting the interactive setup wizard.
Comment on lines
+455
to
+468
| try { | ||
| # Run the setup wizard | ||
| & $pythonCmd -m praisonai setup | ||
| Write-Success "Setup wizard completed successfully!" | ||
| Write-Host "" | ||
| Write-Host "You're all set! 🎉" -ForegroundColor Green | ||
| Write-Host "" | ||
| } catch { | ||
| Write-Warning "Setup wizard failed or was cancelled." | ||
| Write-Host "" | ||
| Write-Host "Don't worry! You can run the setup wizard anytime with:" -ForegroundColor Yellow | ||
| Write-Host " praisonai setup" -ForegroundColor Cyan | ||
| Write-Host "" | ||
| } |
Contributor
There was a problem hiding this comment.
try/catch won't catch a non-zero exit from python -m praisonai setup.
Native-command failures in PowerShell do not raise terminating errors by default, so this try/catch only fires on things like "python.exe not found". If the wizard exits with a non-zero status (user cancels with Ctrl-C, validation error, invalid provider, etc.), the catch block is skipped and the script prints "Setup wizard completed successfully! 🎉" — the opposite of what happened.
Check $LASTEXITCODE explicitly:
Proposed fix
- try {
- # Run the setup wizard
- & $pythonCmd -m praisonai setup
- Write-Success "Setup wizard completed successfully!"
- Write-Host ""
- Write-Host "You're all set! 🎉" -ForegroundColor Green
- Write-Host ""
- } catch {
- Write-Warning "Setup wizard failed or was cancelled."
- Write-Host ""
- Write-Host "Don't worry! You can run the setup wizard anytime with:" -ForegroundColor Yellow
- Write-Host " praisonai setup" -ForegroundColor Cyan
- Write-Host ""
- }
+ try {
+ & $pythonCmd -m praisonai setup
+ $wizardExit = $LASTEXITCODE
+ } catch {
+ $wizardExit = 1
+ }
+
+ if ($wizardExit -eq 0) {
+ Write-Success "Setup wizard completed successfully!"
+ Write-Host ""
+ Write-Host "You're all set! 🎉" -ForegroundColor Green
+ Write-Host ""
+ } else {
+ Write-Warning "Setup wizard failed or was cancelled."
+ Write-Host ""
+ Write-Host "Don't worry! You can run the setup wizard anytime with:" -ForegroundColor Yellow
+ Write-Host " praisonai setup" -ForegroundColor Cyan
+ Write-Host ""
+ }📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| try { | |
| # Run the setup wizard | |
| & $pythonCmd -m praisonai setup | |
| Write-Success "Setup wizard completed successfully!" | |
| Write-Host "" | |
| Write-Host "You're all set! 🎉" -ForegroundColor Green | |
| Write-Host "" | |
| } catch { | |
| Write-Warning "Setup wizard failed or was cancelled." | |
| Write-Host "" | |
| Write-Host "Don't worry! You can run the setup wizard anytime with:" -ForegroundColor Yellow | |
| Write-Host " praisonai setup" -ForegroundColor Cyan | |
| Write-Host "" | |
| } | |
| try { | |
| & $pythonCmd -m praisonai setup | |
| $wizardExit = $LASTEXITCODE | |
| } catch { | |
| $wizardExit = 1 | |
| } | |
| if ($wizardExit -eq 0) { | |
| Write-Success "Setup wizard completed successfully!" | |
| Write-Host "" | |
| Write-Host "You're all set! 🎉" -ForegroundColor Green | |
| Write-Host "" | |
| } else { | |
| Write-Warning "Setup wizard failed or was cancelled." | |
| Write-Host "" | |
| Write-Host "Don't worry! You can run the setup wizard anytime with:" -ForegroundColor Yellow | |
| Write-Host " praisonai setup" -ForegroundColor Cyan | |
| Write-Host "" | |
| } |
🧰 Tools
🪛 PSScriptAnalyzer (1.25.0)
[warning] Missing BOM encoding for non-ASCII encoded file 'install.ps1'
(PSUseBOMForUnicodeEncodedFile)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/scripts/install.ps1` around lines 455 - 468, The try/catch
around running the setup wizard won't catch non-terminating failures from the
native python process; after invoking & $pythonCmd -m praisonai setup, check
$LASTEXITCODE and treat any non-zero value as failure: move the success messages
behind a conditional that verifies $LASTEXITCODE -eq 0 and run the existing
warning/help text when it's non-zero (the same handling currently in the catch
block); keep the catch for true terminating errors (e.g., missing python) but
ensure $pythonCmd and the praisonai setup invocation are followed by an explicit
$LASTEXITCODE check to decide success vs failure.
Comment on lines
+644
to
+655
| if "$py" -m praisonai setup < /dev/tty; then | ||
| log_success "Setup wizard completed successfully!" | ||
| echo "" | ||
| echo -e "${BOLD}${GREEN}You're all set! 🎉${NC}" | ||
| echo "" | ||
| else | ||
| log_warn "Setup wizard failed or was cancelled." | ||
| echo "" | ||
| echo -e "${YELLOW}Don't worry! You can run the setup wizard anytime with:${NC}" | ||
| echo -e " ${CYAN}praisonai setup${NC}" | ||
| echo "" | ||
| fi |
Contributor
There was a problem hiding this comment.
Only stdin is re-attached to the TTY; Rich prompts also need a TTY on stdout/stderr.
When the installer is run via curl … | bash, only stdin is a pipe — stdout/stderr typically stay attached to the terminal, so < /dev/tty is usually sufficient. But if a user does something like curl … | bash > install.log 2>&1, the Rich wizard will detect a non-TTY stdout and silently fall back to non-interactive defaults (or error out), yet run_onboarding will still report "Setup wizard completed successfully!" based on exit code.
Consider either (a) also redirecting stdout/stderr from /dev/tty when appropriate, or (b) explicitly checking [ -t 1 ] in addition to [ -e /dev/tty ] before invoking the wizard.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/scripts/install.sh` around lines 644 - 655, The current
invocation in install.sh that runs the setup wizard (the if "$py" -m praisonai
setup < /dev/tty; then ... in run_onboarding) only re-attaches stdin to /dev/tty
and can still leave stdout/stderr non-TTY (causing Rich prompts to fall back and
a misleading success message); update run_onboarding to either (A) also attach
stdout and stderr to the TTY when launching the wizard (redirect >/dev/tty and
2>/dev/tty along with the existing stdin redirection) or (B) perform an explicit
TTY check before launching (e.g. verify [ -t 0 ] && [ -t 1 ] && [ -t 2 ] or at
minimum [ -t 1 ]) and skip or warn if not all required streams are TTYs so the
wizard is not invoked in a broken/non-interactive environment.
Comment on lines
+8
to
+11
| import os | ||
| import subprocess | ||
| import tempfile | ||
| from pathlib import Path |
Contributor
There was a problem hiding this comment.
Make the subprocess tests independent of the GitHub runner path.
The hard-coded cwd="/home/runner/work/PraisonAI/PraisonAI/src/praisonai" will fail outside that exact checkout location. Use a path derived from __file__, and invoke the current test interpreter.
Proposed fix
import os
import subprocess
+import sys
import tempfile
from pathlib import Path
import pytest
+PROJECT_ROOT = Path(__file__).resolve().parents[2]
+
`@pytest.fixture`
def temp_praison_home():
@@
cmd = [
- "python", "-m", "praisonai", "setup",
+ sys.executable, "-m", "praisonai", "setup",
@@
- cwd="/home/runner/work/PraisonAI/PraisonAI/src/praisonai"
+ cwd=PROJECT_ROOT,
)
@@
- cmd = ["python", "-m", "praisonai", "setup", "--help"]
+ cmd = [sys.executable, "-m", "praisonai", "setup", "--help"]
@@
- cwd="/home/runner/work/PraisonAI/PraisonAI/src/praisonai"
+ cwd=PROJECT_ROOT,
)
@@
cmd = [
- "python", "-m", "praisonai", "setup",
+ sys.executable, "-m", "praisonai", "setup",
@@
- cwd="/home/runner/work/PraisonAI/PraisonAI/src/praisonai"
+ cwd=PROJECT_ROOT,
)
@@
base_cmd = [
- "python", "-m", "praisonai", "setup",
+ sys.executable, "-m", "praisonai", "setup",
@@
- cwd="/home/runner/work/PraisonAI/PraisonAI/src/praisonai"
+ cwd=PROJECT_ROOT,
)
@@
- cwd="/home/runner/work/PraisonAI/PraisonAI/src/praisonai"
+ cwd=PROJECT_ROOT,
)Based on learnings, test files must not depend on timing, external state, or network calls; use mocks and fixtures for isolation; ensure all tests are deterministic.
Also applies to: 33-48, 65-73, 83-97, 121-149
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/tests/integration/test_setup_integration.py` around lines 8 -
11, Tests currently use a hard-coded GitHub runner path and a plain "python"
invocation; change subprocess calls to use sys.executable and compute the
working directory relative to the test file using Path(__file__).resolve()
(e.g., package_dir = Path(__file__).resolve().parents[1] or appropriate parent
to reach the src/praisonai package) and pass cwd=str(package_dir) to
subprocess.run/ Popen; update all occurrences referenced in the file (lines
~33-48, 65-73, 83-97, 121-149) and replace "python" with sys.executable to
ensure the current test interpreter is used, and where external state, timing or
network calls are involved wrap/replace them with mocks/fixtures (use
unittest.mock or pytest fixtures) so the subprocess tests are deterministic and
runner-path independent.
Comment on lines
+69
to
+70
| # Check file permissions (should be 600) | ||
| assert oct(env_file.stat().st_mode)[-3:] == "600" |
Contributor
There was a problem hiding this comment.
Permission assertion will fail on Windows.
os.chmod(path, 0o600) on Windows only toggles the read-only bit, so oct(stat_mode)[-3:] typically evaluates to "666" / "444", not "600". If this test suite is intended to run on Windows CI, guard these assertions with pytest.mark.skipif(os.name == "nt", ...) or check sys.platform.
Also applies to: 216-230
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/praisonai/tests/unit/cli/test_setup_command.py` around lines 69 - 70, The
permission assertion using oct(env_file.stat().st_mode)[-3:] == "600" will fail
on Windows; update the test(s) that check file permissions (the assertion around
env_file.stat().st_mode and the similar block at lines 216-230) to skip on
Windows using pytest.mark.skipif(os.name == "nt" or
sys.platform.startswith("win"), reason="permission bits not reliable on
Windows") or alternatively replace the exact-mode assertion with a
platform-aware check (e.g., only assert presence of owner read/write on POSIX
via os.name == "posix" guard before checking st_mode). Ensure you reference the
existing assertion expression env_file.stat().st_mode when locating and
modifying the tests.
Owner
|
@claude You are the FINAL architecture reviewer. If the branch is under MervinPraison/PraisonAI (not a fork), you are able to make modifications to this branch and push directly. SCOPE: Focus ONLY on Python packages (praisonaiagents, praisonai). Do NOT modify praisonai-rust or praisonai-ts. Read ALL comments above from Gemini, Qodo, CodeRabbit, and Copilot carefully before responding. Phase 1: Review per AGENTS.md
Phase 2: FIX Valid Issues Phase 3: Final Verdict |
Contributor
Author
|
Claude finished @MervinPraison's task —— View job Final Architecture Review & Fix Implementation ✅ COMPLETEPhase 1: Architecture Review per AGENTS.md
Phase 2: Fix Valid Issues from Reviewers
Phase 3: Final Verdict
Architecture Review Results✅ PROTOCOL-DRIVEN: Setup feature correctly uses optional dependencies and lazy imports Issues Fixed & Committed (SHA: 00806cd)Critical Security & Architecture Issues:
Platform Compatibility & Robustness:
Final Verdict: ✅ APPROVEDThis PR successfully implements frictionless onboarding following AGENTS.md principles:
The setup wizard provides excellent developer experience while maintaining the framework's architectural integrity. All critical issues from reviewers have been resolved. Recommendation: Merge this PR. The setup feature will significantly improve user onboarding experience. |
- Add remediation hints to PraisonAIConfigError for better guidance - Fix setup reset to remove both .env and config.yaml files - Create directories with secure 700 permissions - Enhance non-interactive validation (prefer env vars, require model) - Use atomic .env file creation with secure permissions - Fix starter YAML to use selected model instead of hardcoded gpt-4o-mini - Add TTY/interactive checks to PowerShell installer - Improve shell script python command precedence - Remove unnecessary f-string prefixes (ruff F541) - Make tests platform-independent and Windows-compatible - Export PraisonAIConfigError in errors.__all__ Addresses CodeRabbit security and architecture findings. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Mervin Praison <MervinPraison@users.noreply.github.com>
There was a problem hiding this comment.
praisonai-triage-agent[bot] has reached the 50-review limit for trial accounts. To continue receiving code reviews, upgrade your plan.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Implements comprehensive frictionless onboarding system per Issue #1451
Key Features
Implementation Details
Files Changed
🤖 Generated with Claude Code
Summary by CodeRabbit
New Features
praisonai setupcommand for onboarding and provider configurationwizard(interactive),config(view/edit settings), andreset--no-onboardflag)Bug Fixes
praisonai setupfor API key configurationTests