> ## Documentation Index > Fetch the complete documentation index at: https://praison.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # PraisonAI Code CLI > Standalone agentic terminal CLI: run agents, chat, and code without the full wrapper > `praisonai-code` is the terminal-native agent CLI — install it on its own for a smaller footprint when you only need agentic commands. ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Standalone install — no wrapper needed from praisonaiagents import Agent Agent(instructions="Summarise this URL").start("https://arxiv.org/abs/2409.12345") ``` ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Same result from the CLI, standalone (PR #2853) pip install praisonai-code praisonai-code run "Summarise https://arxiv.org/abs/2409.12345" ``` The user runs a terminal prompt; `praisonai-code` executes the agent in-process and prints the reply. Interactive `chat` and `code` also run from `praisonai-code` alone — no wrapper (PR #3818): ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai-code praisonai chat # opens the split-pane async TUI, no wrapper required praisonai code # opens the resident code TUI, no wrapper required ``` ```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} graph LR subgraph pip_code["pip install praisonai-code"] A[praisonai-code CLI] --> A1[default run · run --output plain/verbose/silent/actions/json/stream/stream-json · config · doctor · daemon · version · …] A --> A2[chat · code · resident TUI] end subgraph pip_wrapper["pip install praisonai"] B[praisonai CLI] --> B1[gateway · bot · onboard · pairing · identity · kanban · claw · dashboard] B --> A end A --> C[praisonaiagents core] classDef code fill:#189AB4,stroke:#7C90A0,color:#fff classDef wrapper fill:#8B0000,stroke:#7C90A0,color:#fff classDef core fill:#10B981,stroke:#7C90A0,color:#fff class A,A1,A2 code class B,B1 wrapper class C core ``` As of [PR #3818](https://github.com/MervinPraison/PraisonAI/pull/3818), `chat` and `code` interactive sessions now run **resident inside `praisonai-code`** — no wrapper required. `pip install praisonai-code` alone delivers a full interactive `chat`/`code` session on the split-pane TUI. Default `run "…"` and every `--output` text mode already ran in-process (PR #2853). When the `praisonai` wrapper **is** installed, it is still detected and the richer legacy interactive dispatch is used; the resident TUI is the fallback when the wrapper is absent. ## Which install do I need? ```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} graph TB Start{What do you run?} -->|Any text prompt: default run or --output plain/verbose/silent/actions/json/stream/stream-json| Structured[run "…"] Start -->|Interactive terminal chat or coding: resident TUI| Interactive[chat · code] Start -->|Bots · gateway · dashboard| Wrapper3[gateway · bot · …] Start -->|Warm runtime daemon| Daemon[daemon start] Structured --> Code[pip install praisonai-code] Interactive --> Code Daemon --> Code Wrapper3 --> Full[pip install praisonai] classDef question fill:#6366F1,stroke:#7C90A0,color:#fff classDef code fill:#189AB4,stroke:#7C90A0,color:#fff classDef wrapper fill:#8B0000,stroke:#7C90A0,color:#fff classDef install fill:#10B981,stroke:#7C90A0,color:#fff class Start question class Structured,Interactive,Daemon code class Wrapper3 wrapper class Code,Full install ``` Both `praisonai-code` (console script) and `python -m praisonai_code` call the same entry point: configure logging, register commands, then run the Typer app. ## Interactive Chat & Code (resident TUI) A single `pip install praisonai-code` is enough for an interactive session — the split-pane async TUI ships inside the package (PR #3818). ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai-code praisonai chat # opens the split-pane async TUI, no wrapper required praisonai code # opens the resident code TUI, no wrapper required ``` ```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} graph LR subgraph "Interactive session" I[💻 praisonai chat / code] --> C{🔍 wrapper installed?} C -->|No| R[🖥️ Resident TUI
praisonai_code.cli.interactive] C -->|Yes| L[🎛️ Richer legacy dispatch
PraisonAI._start_interactive_mode] end classDef input fill:#6366F1,stroke:#7C90A0,color:#fff classDef decision fill:#F59E0B,stroke:#7C90A0,color:#fff classDef result fill:#10B981,stroke:#7C90A0,color:#fff class I input class C decision class R,L result ``` | Dimension | Resident TUI (`praisonai-code` alone) | Richer legacy dispatch (`+ praisonai`) | | ---------------------------------------- | -------------------------------------------------------------- | --------------------------------------------------------- | | Install | `pip install praisonai-code` | `pip install praisonai` | | Entry point | `praisonai_code.cli.interactive` | `PraisonAI._start_interactive_mode` | | `code` command | Resident split-pane TUI | Preferred richer legacy dispatch | | Instruction loading (`context_files`) | Native — `praisonai_code.integration.context_files` (PR #4338) | Prefers `praisonai_bot` / `praisonai` copy when installed | | Async bridge (`praisonai._async_bridge`) | Bridge fallback via `_wrapper_bridge` | Direct wrapper implementation | The wrapper is never required for the baseline TUI — it only upgrades `code` to the richer legacy dispatch. Instruction loading is now resident: `praisonai_code.integration.context_files` ships inside the package, so the subtree hook and `--instructions` globs/URLs work standalone without routing through `_wrapper_bridge`. When the wrapper (or `praisonai_bot`) is installed, the CLI prefers that copy. ### Native rules + project instructions `rules list/add/clear` and subtree instruction loading run natively on a standalone `pip install praisonai-code` (PR #4338) — no `praisonai_bot` or `praisonai` needed. ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai-code praisonai rules add my_rule "Always use type hints" # workspace-scoped rule praisonai rules list # native, backed by RulesManager praisonai rules clear # clears workspace rules only ``` * `rules list/add/clear` run natively via `praisonaiagents.memory.RulesManager`. * Subtree `AGENTS.md`/`CLAUDE.md` auto-attach and `--instructions` globs/URLs both work standalone via `praisonai_code.integration.context_files`. * `clear` is scoped to **workspace rules only** — hand-authored `AGENTS.md`/`CLAUDE.md`/`PRAISON.md`, global rules, git-root rules, and nested subdir rules are never removed. ```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} graph TB subgraph "Standalone praisonai-code project-instruction stack" USER[👤 User] --> CLI[💻 praisonai-code run / chat / code] CLI --> INSTR[🔍 --instructions globs/URLs
+ AGENTS.md walk-up] CLI --> RULES[📜 rules list/add/clear] INSTR --> RESIDENT[📦 praisonai_code.integration.context_files
resident, stdlib + praisonaiagents.hooks] RULES --> RM[🧠 praisonaiagents.memory.RulesManager
PRAISON.md / CLAUDE.md / AGENTS.md / .praison/rules/] RESIDENT --> AGENT[🤖 Agent] RM --> AGENT end classDef user fill:#6366F1,stroke:#7C90A0,color:#fff classDef cli fill:#8B0000,stroke:#7C90A0,color:#fff classDef process fill:#F59E0B,stroke:#7C90A0,color:#fff classDef native fill:#189AB4,stroke:#7C90A0,color:#fff classDef agent fill:#10B981,stroke:#7C90A0,color:#fff class USER user class CLI cli class INSTR,RULES process class RESIDENT,RM native class AGENT agent ``` Deep dives: [Rules](/docs/features/rules), [Instruction Sources](/docs/features/instruction-sources), and [Context Files](/docs/features/context-files). ## Quick Start ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai-code praisonai-code version ``` The panel lists **PraisonAI Code**, **PraisonAI Agents**, and **Python**. When the full wrapper is installed, a **PraisonAI Wrapper** line appears as well. Use `--version` for a quick one-liner (package version only, no panel): ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} praisonai-code --version ``` Run your first agent: ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} praisonai-code run "Summarise the top 3 arXiv papers on RAG this week" ``` Every text mode — default `run "…"` and `--output plain|verbose|silent|actions|json|stream|stream-json` — runs standalone via the in-process `Agent` path. As of PR #3818, `chat` and `code` also open their resident interactive TUI standalone — the wrapper adds only the richer legacy dispatch for `code` and the gateway/bot channels. ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai ``` The same `praisonai-code` binary keeps working. Wrapper-only commands (`gateway`, `bot`, `onboard`, `pairing`, `identity`, `kanban`, `claw`, `dashboard`) become available through the composed install. ## How It Works Text-mode `run` always executes in-process via the `Agent` path on a standalone install. When the wrapper **is** present, human-readable modes delegate to its `handle_direct_prompt`; when it is absent, they route through the in-process `Agent` with the matching output preset. `chat` and `code` open their resident TUI standalone (PR #3818); when the wrapper is installed, `code` prefers the richer legacy dispatch (`PraisonAI._start_interactive_mode`). Plugin discovery is vendored inside `praisonai-code`, so plugin-related flows keep working without the wrapper. ```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} sequenceDiagram participant User participant CLI as praisonai-code participant Bridge as _wrapper_bridge participant Agent as praisonaiagents.Agent participant Wrapper as praisonai (optional) User->>CLI: praisonai-code run "…" CLI->>Bridge: wrapper_available() alt wrapper installed Bridge-->>CLI: True CLI->>Wrapper: handle_direct_prompt Wrapper-->>CLI: result CLI-->>User: normal output else wrapper missing — default / plain / verbose / silent / actions / json / stream / stream-json Bridge-->>CLI: False CLI->>Agent: in-process run (output preset) Agent-->>CLI: result CLI-->>User: final text (silent-style presets print the answer) end Note over User,CLI: chat / code run resident standalone — wrapper only adds richer legacy dispatch ``` Output presets map as `plain` → `plain`, `silent`/default → `silent`, `verbose` → `verbose`; silent-style presets print the final text. `chat` and `code` open their resident interactive TUI standalone (PR #3818) — `pip install praisonai-code` alone delivers full interactive sessions. When the `praisonai` wrapper is installed, `code` prefers its richer legacy dispatch (`PraisonAI._start_interactive_mode`): ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai-code praisonai-code chat praisonai-code code "Refactor this function" ``` The vendored `_registry` module means plugins work even without the wrapper installed. Version resolution reads from the `praisonai-code` package metadata directly, not the wrapper. ## Command matrix Three tiers decide what a command needs: fully standalone, wrapper-boosted (works standalone but gains a richer path with the wrapper), and wrapper-only (not listed without the wrapper). | Standalone (`pip install praisonai-code`) | Wrapper-boosted (works standalone, richer with `praisonai`) | Wrapper-only (`pip install praisonai`) | | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------- | --------------------------------------------------------------------------------- | | acp · agent · agents · attach · auth · batch · benchmark · browser · call · `chat` · checkpoint · `code` · command · commit · completion · config · context · debug · deploy · diag · docs · doctor · endpoints · env · eval · examples · flow · github · hooks · init · knowledge · langextract · langfuse · loop · lsp · managed · mcp · memory · models · n8n · obs · package · paths · permissions · plugins · port · profile · publish · rag · realtime · recipe · registry · replay · research · `run "…"` (default) · `run --output plain/verbose/silent/actions/json/stream/stream-json` · rules · sandbox · schedule · serve · session · setup · skills · templates · test · todo · tools · traces · tracker · train · ui · up · validate · version · workflow | code (resident TUI standalone → richer legacy dispatch with wrapper) | bot · claw · daemon · dashboard · gateway · identity · kanban · onboard · pairing | `chat` and `code` open their resident TUI standalone (PR #3818); `code` prefers the richer legacy dispatch when the wrapper is installed. Wrapper-only names stay out of `--help` entirely and do not load a Typer module. The 13 stub groups below are a special mixed case — their legacy entry points are wrapper-required, but native subcommands stay standalone-safe. `rules` is **not** in this list — `rules list/add/clear` are native standalone (PR #4338): | Command | Standalone | Wrapper-required | | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :-------------------: | :--------------: | | Stub groups (`agents`, `workflow`, `registry`, `memory`, `skills`, `hooks`, `eval`, `package`, `templates`, `todo`, `research`, `commit`, `call`) — delegating entry points | ❌ Fail fast with hint | ✅ | | Stub groups — native subcommands (`memory learn`, `skills bundle/check/eligible`, `eval list-judges`/`list`) | ✅ | ✅ | `daemon start` runs standalone in both foreground and `--background` modes — only the bot/gateway daemon sub-apps are wrapper-only. ### Standalone limits On a `pip install praisonai-code`-only install: | Command | Works standalone? | Notes | | ------------------------------------------------------- | ----------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `run --help`, `config`, `doctor` | Yes | | | `run --output actions "…"` | Yes | In-process `Agent` (structured events preset) | | `run --output json "…"` | Yes | In-process `Agent` (structured JSON output) | | `run --output stream "…"` | Yes | In-process `Agent` (streaming text) | | `run --output stream-json "…"` | Yes | In-process `Agent` (streaming JSON events) | | `run "…"` (default) | Yes | In-process `Agent` (silent preset; prints final text) | | `run --output plain/verbose/silent "…"` | Yes | In-process `Agent` (plain/verbose/silent presets) | | `chat`, `code` | Yes | Run resident on the built-in split-pane TUI (PR #3818); the `praisonai` wrapper adds a richer dispatch mode for `code` | | `daemon start` (foreground) | Yes | | | `daemon start --background` | Yes | Spawns `python -m praisonai_code.runtime` | | Piped stdin (`type file.log \| praisonai-code run "…"`) | Yes | Works on Windows as of 2026-07-07 (PR #2705) via a stat-based pipe classifier; also supports `Get-Content file.log \| praisonai-code run "…"`. Interactive terminals skip the stdin read. See [Piped Input](/docs/docs/features/cli-piped-stdin#size-cap--platform-notes). | `chat`, `code`, and every `run` output mode work on a `praisonai-code`-only install (PR #2853, PR #3818). Installing the wrapper (`pip install praisonai`) adds the richer legacy dispatch for `code`, the gateway, and channel bots — it is no longer required for the baseline interactive experience. `praisonai-code doctor` returns exit **0** on a healthy standalone install — wrapper-presence checks (`performance_praisonai_import`, `acp_module`, `praisonai_package_structure`, `console_script_execution`) SKIP rather than FAIL ([PR #2851](https://github.com/MervinPraison/PraisonAI/pull/2851)). ### Stub command groups on standalone Run agents standalone without the wrapper — the SDK works directly: ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} from praisonaiagents import Agent Agent(instructions="Summarise this URL").start("https://arxiv.org/abs/2409.12345") ``` ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Same result from the CLI, standalone (in-process) pip install praisonai-code praisonai-code run --output actions "Summarise https://arxiv.org/abs/2409.12345" ``` Thirteen command groups — `agents`, `workflow`, `registry`, `memory`, `skills`, `hooks`, `eval`, `package`, `templates`, `todo`, `research`, `commit`, and `call` — delegate to the `praisonai` wrapper for their legacy entry points. `rules` is not among them — `rules list/add/clear` run natively on standalone (PR #4338). On a standalone `pip install praisonai-code` ([PR #2854](https://github.com/MervinPraison/PraisonAI/pull/2854)) those entry points exit `1` with a single-line install hint — no Rich traceback: ``` agents requires the full wrapper. Install the full wrapper: pip install praisonai ``` The `` name changes per command (`workflow requires the full wrapper. …`, `registry requires the full wrapper. …`, and so on). These groups are **mixed**, not wholly wrapper-only. Their native subcommands keep working standalone — only the legacy delegating entry points fail fast. | Group | Standalone-safe native subcommand(s) | | -------- | ---------------------------------------------------------------------------------------------------------- | | `memory` | `memory learn` (status/show/add/search/clear) | | `skills` | `skills bundle`, `skills check`, `skills eligible` | | `eval` | `eval list-judges`, `eval list` (alias — PR [#2848](https://github.com/MervinPraison/PraisonAI/pull/2848)) | Run `pip install praisonai` to unlock the wrapper-delegating entry points, or use the native standalone-safe subcommand. ```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} graph TB subgraph "praisonai-code stub group on standalone" Cmd[User: praisonai-code group sub] --> Kind{Subcommand kind?} Kind -->|Native: memory learn · skills bundle/check/eligible · eval list-judges/list| InProc[Runs in-process] Kind -->|Delegating stub: agents list · workflow run · registry add| Guard{Wrapper installed?} Guard -->|Yes| Wrap[Delegates to praisonai wrapper] Guard -->|No| Hint[exit 1 with single-line install hint] end classDef input fill:#6366F1,stroke:#7C90A0,color:#fff classDef gate fill:#F59E0B,stroke:#7C90A0,color:#fff classDef ok fill:#10B981,stroke:#7C90A0,color:#fff classDef fail fill:#8B0000,stroke:#7C90A0,color:#fff class Cmd,Kind input class Guard gate class InProc,Wrap ok class Hint fail ``` ### Wrapper-command loading Wrapper-only commands (`bot`, `gateway`, `pairing`, `identity`, `onboard`, `kanban`, `dashboard`, `claw`, `daemon`) are Typer sub-apps whose implementations live in the `praisonai` wrapper. When you install `praisonai-code` standalone, these commands are not listed in `--help` and are not resolvable — `get_command()` returns `None` instead of loading a module that doesn't exist in `praisonai_code`. ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Standalone install, wrapper-only command $ pip install praisonai-code $ praisonai-code bot --help Error: No such command 'bot'. ``` To use `bot`, `gateway`, `pairing`, `identity`, `onboard`, `kanban`, `dashboard`, `claw`, or `daemon`, install the full wrapper: ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} pip install praisonai ``` ## Configuration and environment | Variable / command | Behaviour | | ------------------------------ | ------------------------------------------------------------------------------------------------------------------- | | `LOGLEVEL` | Read on CLI entry via `configure_cli_logging` (default `WARNING`). Controls root log verbosity for standalone runs. | | `praisonai-code version check` | Compares your installed version with PyPI (`https://pypi.org/pypi/praisonai-code/json`). | ### Version commands ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} praisonai-code version ``` Shows the full version panel: ``` PraisonAI Code: 0.0.4 PraisonAI Wrapper: 1.x.x ← only shown when praisonai is installed PraisonAI Agents: 1.6.x Python: 3.12.x ``` ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} praisonai-code version --json ``` Returns structured JSON: ```json theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} { "praisonai-code": "0.0.4", "praisonai": "1.x.x", "praisonaiagents": "1.6.x", "python": "3.12.x" } ``` The `praisonai` key is omitted when the wrapper is not installed. ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} praisonai-code version check ``` Queries PyPI and returns update status: ```json theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} { "current": "0.0.4", "latest": "0.0.5", "update_available": true } ``` `praisonai-code version check` needs outbound HTTPS to PyPI. ## Best Practices Use `praisonai-code` for any `run "…"` prompt (default or any `--output` mode), interactive `chat`/`code` (resident TUI, PR #3818), `config`, `doctor`, or the warm-runtime daemon — smaller install, no gateway/bot deps. Install `praisonai` for the richer legacy `code` dispatch (`PraisonAI._start_interactive_mode`), Telegram/Discord/Slack bots, the WebSocket gateway, `kanban`/`dashboard`, and `RunPolicy` for unattended runs. Baseline interactive `chat`/`code` run standalone (PR #3818). In dev, `pip install -e src/praisonai-agents && pip install -e src/praisonai-code && pip install -e src/praisonai` (this exact order) — matches the release publish order in `pypi-release.yml`. `from praisonai import AgentOS` and `from praisonai import AgentApp` both work — `AgentApp` is a backward-compat silent alias for `AgentOS`. No deprecation warning is emitted. ## Related Compare full wrapper, code-only, and SDK installs Six-package layout and dependency direction