Skip to content

getting-started-solo-cli.md

Latest commit

 

History

History
46 lines (33 loc) · 1.52 KB

getting-started-solo-cli.md

File metadata and controls

46 lines (33 loc) · 1.52 KB
Raw

Getting Started — Solo CLI User

Last reviewed: 2026-06-06 Persona: Individual developer running TeaAgent locally from the terminal

You want a governed terminal agent — not an IDE plugin — with clear permissions, cost caps, and an audit trail you can inspect after each run.

Prerequisites

  • Python 3.11+ (see README for venv setup)
  • A model provider API key (OpenAI, Anthropic, etc.)

15-minute path

pip install -e .
teaagent setup --root . --provider gpt --permission-mode read-only --write-env
source ~/.teaagent/providers_env.zsh   # or your shell snippet

teaagent daily "what should I work on?" --human --root .
teaagent run "summarize the test suite" --permission-mode read-only --root .

Daily habits

Habit Command
Read-only exploration --permission-mode read-only
Writes with approval --permission-mode prompt
Cost cap --max-estimated-cost-cents 500
See what happened teaagent agent runs show <run_id> --receipt --root .
Undo last write teaagent agent undo --last --root .

When to escalate modes

  1. Plan / explore → read-only
  2. Implement with guardrails → workspace-write (plan required by default)
  3. Interactive destructive → prompt
  4. Trusted automation only → allow or danger-full-access

Next steps