agentleFS
AlbertoVasquezRAlbertoVasquezR/panal/llms.txt

panal

llms.txt1 starsChanged 45 days agoApache-2.0View on GitHub

Panal (panal) is a terminal dashboard, written in Go with Bubble Tea, for the AI coding agents orchestrated from Claude Code: opencode, codex, agy and Claude Code itself. It shows live cards with pixel-art mascots, live quotas, run history, reports and a timeline, and only reads the agents' files. panal delegate hands a task to codex, agy or opencode with quota fallback, and its router (-c auto) learns from every run which agent, model and effort fits each task type…

  • Installs packages

The file

Files
llms.txt
# Panal

> Panal (`panal`) is a terminal dashboard, written in Go with Bubble Tea, for the AI coding agents orchestrated from Claude Code: opencode, codex, agy and Claude Code itself. It shows live cards with pixel-art mascots, live quotas, run history, reports and a timeline, and only reads the agents' files. `panal delegate` hands a task to codex, agy or opencode with quota fallback, and its router (`-c auto`) learns from every run which agent, model and effort fits each task type and tier (Beta posteriors, Thompson sampling, cost-aware), with `panal route` as a dry run and `panal feedback` to correct it.

Module `github.com/AlbertoVasquezR/panal`, Apache-2.0 license, Go 1.27+. Install with `go install github.com/AlbertoVasquezR/panal/cmd/panal@latest`. Panal writes only its own files in `~/.panal`.

## Docs

- [README](README.md): overview, quick start, commands, keys
- [Getting started](docs/getting-started.md): install, connecting each agent, the Claude Code status line, `panal -doctor`
- [Dashboard](docs/dashboard.md): views, cards, every key, glyphs, mascots, live quota, alerts, themes
- [Pet](docs/pet.md): `panal pet`, one mascot for a small split pane next to an agent CLI; `-json` / `-stream` frames; the `panal-pet` Claude Code plugin in `plugins/panal-pet`
- [Delegating](docs/delegate.md): `panal delegate` flags, chain, exit codes, sandboxing per agent, files written
- [Router](docs/router.md): tier rules, outcomes, priors and formulas, Thompson sampling, feedback, `panal route`, external router contract
- [Models](docs/models.md): model discovery per CLI, cache, `pool = auto`, `models` / `exclude` patterns, cost heuristic
- [Configuration](docs/configuration.md): every `panal.conf` key and environment variable
- [Reference](docs/reference.md): run file JSON schema, report formats, status line, `-serve` for multiple machines, the `panal pet` JSON frame contract

## Contributing

- [AGENTS.md](AGENTS.md): rules, commands, repo map, invariants, where to change what
- [Architecture](docs/ARCHITECTURE.md): data flow, package responsibilities, key types
- [CONTRIBUTING.md](CONTRIBUTING.md): build, golden screens, ground rules

## Source

- [cmd/panal](cmd/panal/main.go): flags and subcommand dispatch
- [internal/router](internal/router): tiers, outcomes, posteriors, Thompson choice
- [internal/delegate](internal/delegate): `panal delegate`, `route`, `models`
- [internal/models](internal/models): model discovery and cost heuristic
- [internal/readers](internal/readers): one read-only reader per agent
- [internal/runs](internal/runs/runs.go): the run file format
- [internal/ui](internal/ui): the Bubble Tea dashboard
- [internal/pet](internal/pet): `panal pet`, its terminal view and JSON frames

## Optional

- [Work journal](docs/JOURNAL.md): what was done, why, and what is pending
- [UX audit](docs/ux-audit.md): open UI findings
- [UI design guide](.claude/skills/panal-ux/SKILL.md): palette, glyphs, mascots, sizes

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.