CLI command reference
Every command, flag, and REPL shortcut in tempr — plus the tool set the agent drives. New here? Start with the CLI quickstart.
Commands
| Command | Does |
|---|---|
tempr "<prompt>" | One-shot agent turn: a real tool-calling loop that reads, edits, runs commands, and searches your codebase, then exits. |
tempr | Interactive REPL with persistent cross-session history, line editing, and tab-completion for /commands and @personas. |
<something> | tempr | Headless/piped mode — reads stdin as the prompt context, writes to stdout. See Scripts, pipes & CI. |
tempr auth | Sign in with your Tempr license key (the same license as the IDE). |
tempr logout | Clear the stored credential on this machine. |
tempr config | Manage settings — tempr config set-model <id> sets your default model. |
tempr history | List saved sessions; resume one with tempr --id <SESSION_ID>. |
tempr mcp | Add, list, and manage MCP servers for the terminal agent. |
tempr doctor | Check connectivity and diagnose your setup. |
tempr --version | Print the installed CLI version. |
Options
These flags apply to a chat or agent run:
| Flag | Does |
|---|---|
-m, --model <MODEL> | Use this model for the run, overriding tempr config set-model. |
-y, --yolo | Auto-approve every tool call for this run — approval-mode bypass for a single invocation. Ideal for scripted, non-interactive runs you trust. |
--ask | Plain single-turn chat with no tool-calling — quick questions, no file access. |
--id <SESSION_ID> | Resume a saved session (see tempr history). A new id is generated and saved automatically when omitted. |
--json | Emit newline-delimited JSON (one event per line) instead of formatted text — built for scripting. See the JSON event stream. |
--verbose | Print token usage (prompt/completion/total, and cost if known) after each turn. |
REPL slash commands
Inside an interactive tempr session:
| Command | Does |
|---|---|
/help | List available commands. |
/model | Show or switch the active model. |
/persona | Show or switch the active persona (also type @name with tab-completion). |
/mcp | Show connected MCP servers and their tools. |
/history | Browse earlier turns in this session. |
/clear | Clear the screen. |
The agent tool set
In agent mode, the model drives the same tool families as the IDE — executed locally against your workspace:
| Tool family | What it does |
|---|---|
| Files | Read, write, and edit files (read first, then targeted edits). |
run_command | Run shell commands, gated by the command-safety whitelist. |
| Semantic search | Embedding-based search over the workspace — finds conceptually related code. |
| Symbols | Look up and navigate code symbols. |
| Tests | Discover and run tests with structured pass/fail results. |
| Diagnostics | Read compiler and analyzer diagnostics. |
| Workspace graph | Project references and call-graph context. |
| Sub-agent delegation | Hand a focused sub-task to a specialist persona mid-turn. |
| MCP bridge | Expose tools from your configured MCP servers. |
Tip
The agent loop runs server-side like the IDE, so a dropped connection doesn't lose an in-flight turn — and every session is saved automatically, ready to resume with tempr --id.
Next steps
- Scripts, pipes & CI — headless patterns, the JSON event stream, and automation examples.
- Agent modes & personas — the persona lineup and how delegation works.
- MCP servers — connect your own tools and data sources.