Bonfire

Your Claude Code, opinionated.

Pip-install a runtime and a Claude Code skill. Run /bonfire scan in chat. Name your assistant — your cadre wires itself around the name.

PyPIbonfire-ai 1.0.1 Python≥ 3.12 LicenseApache-2.0
§01 Install · 60 seconds three steps

Two commands, then chat.

Step 01 · pypi

Install the runtime

One Python package. No daemon, no service, no auth dance.

Step 02 · skill

Register the skill

Drops a Claude Code skill into ~/.claude/skills/bonfire.

$ pip install bonfire-ai          # PyPI distribution; Python 3.12+
$ bonfire install-skill           # registers ~/.claude/skills/bonfire/SKILL.md
$ export ANTHROPIC_API_KEY=sk-ant-...

Then, in Claude Code, in any project, run /bonfire scan — the cadre wires itself around your repo.

> /bonfire scan
> what do you want to call me?
§02 The cadre nine roles · one batch

Nine roles, one batch.

Bonfire wires nine roles around your project. Each plays its role; none impersonate the others. The config file is checked into your repo as bonfire.toml.

01 Scout investigates
02 Knight writes tests
03 Warrior implements
04 Cleric verifies
05 Bard publishes
06 Wizard reviews
07 Steward closes
08 Sage synthesizes
09 Architect analyzes

See the pipeline: How it works → · Seven in-pipeline roles run in fixed order. Sage runs cross-pipeline; Architect runs as a pre-pipeline bracket. Reference: Pipeline stages →

§03 Persona · the naming gesture 3 built-ins

You name your assistant.

The naming gesture.

You name your assistant; the cadre wires itself around the name. Three built-in personas; or define your own. The persona affects voice, not behavior — a Warrior is still a Warrior.

default minimal falcor your-name-here
$ bonfire persona list
falcor    (default)
default
minimal
$ bonfire persona set falcor
§04 · Why it exists

Most agents vibe-code. Bonfire bakes.

structural TDD
role-boundary contracts
audit-driven release gates
per-run cost ledger

Bonfire packages the discipline of a real engineering studio: structural TDD at the role boundary, audit-driven release gates, cost ledgers per run.

Read about the discipline: /bonfire/docs/discipline.

§05 The promise canonical · verbatim

Four claims.

01

It learns.

Vault memory. Priority truncation. Tokens compound because the system remembers what worked last time and what didn't.

02

It competes with itself.

Dual and triple workflows race. Scouts take different lenses on the same ticket. Best output wins. Never a single-shot answer.

03

It feeds its own fire.

Every resolved ticket feeds the vault. The line heats itself. Session N+1 starts warmer than session N because fuel accumulates.

04

It gets better the more you use it.

XP progression across six tiers. The system's handling of your codebase sharpens with volume. Not a plateau tool.

§06 Documentation five docs · one template

Five curated docs.

§07 Open source · provenance apache-2.0

Apache-2.0 · public.

Built and maintained by CandyFactory.

Bonfire is the methodology this studio uses on its own work, packaged. Issues and PRs welcome on github.com/BonfireAI/bonfire. Releases follow semver; ADRs are public.

Read about the working discipline, the nine-role cadre, and the cost ledger.