Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ohm41321/luciazero --skill lucia-chatgit clone --depth 1 https://github.com/ohm41321/luciazeroWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ohm41321/luciazero/lucia-chat)<a href="https://agentmods.dev/skills/ohm41321/luciazero/lucia-chat"><img src="https://agentmods.dev/badge/skills/ohm41321/luciazero/lucia-chat/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ohm41321/luciazero/lucia-chat"><img src="https://agentmods.dev/badge/skills/ohm41321/luciazero/lucia-chat.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00055 | $0.01652 |
| Opus 5 | $0.00028 | $0.00826 |
| Sonnet 5 | $0.00011 | $0.00330 |
| Haiku 4.5 | $0.00006 | $0.00165 |
Grade A, and why
lucia-chat scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lucia Chat
/lucia-bus is how you take part in the bus. This is how the user sets up a
conversation between two other sessions: one window per agent, and an optional
read-only pane showing every message as it lands.
After ./install.sh, lucia is that command from anywhere and
luciazero-agentd is the same program under its long name. Without the
launcher, each command below is the same one run as python3 -m luciazero_agentd from the repository's agentd/ directory, which is the form
the bus itself prints. Run them in the user's terminal, or hand them over to
paste. Never start a provider session on their behalf without being asked to.
1. Start here, always
lucia next
It reads the bus and answers the only question the user actually has — what is waiting on whom — as the command that unblocks it, most blocking first: a delivery nobody could deliver or a task that ran out of budget (both need a person to decide something), then each agent with unread messages and the command that opens its session in its own worktree. If the daemon is down it says so and nothing else, because nothing else can happen first:
lucia serve
Read the answer back to the user and offer to run the command it names; do not
paraphrase it into different commands. status is still there for the full
picture, and next never writes anything.
2. Open one window per agent
lucia claude
lucia codex
The provider is the verb: each starts that provider with the binding already in
place, as the agent whose id is the provider's own name. A second window for the
same provider needs a name of its own or the bus refuses the duplicate —
lucia claude --as reviewer is the agent claude-reviewer. An id that is not a
provider's name takes the long form, which is what chat prints:
lucia run --agent codex-architect -- codex
Each agent's git checkout must be its own: start the second session from a
separate worktree, or worktree_bind refuses it. A daemon is started if this
state directory has none; --no-autostart refuses instead of starting one.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed · +37 lines d47f4ff5cdb3
- 3d ago First seen · 116 lines · 55 tokens per session scan A cfe730a5adec
lucia-chat is a skill published in the GitHub repository ohm41321/luciazero (5 stars, last pushed today), licensed MIT. It adds 55 tokens to every session and 1,652 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other skills, from other repositories
tfx-goal-clarify
A prompt-formatting helper that turns a rough idea written in everyday language into one Claude Code `/goal` instruction. It organizes the desired result, how to check it, limits, and when to stop.
tfx-ship
A release workflow for publishing triflux versions through GitHub Releases, the npm package registry, and the Claude Code marketplace. It also supports version changes, dry runs, skipped tests, and no-publish runs.
tfx-interview
A question-driven workflow for turning a vague implementation request into specific requirements. It measures ambiguity—how much the goal, constraints, and success criteria are unclear—and asks questions to reduce it.
star-prompt
A setup change that adds a GitHub star request to a command-line tool's installation flow. GitHub stars are public bookmarks users can give to repositories on GitHub.
tfx-hub
A Korean-language skill for managing a TFX message hub, which is a local message bus for communication between coding agents. It supports starting, stopping, checking, and routing work through the hub.
tfx-prune
A cleanup workflow that asks three AI coding assistants to independently identify unnecessary code, then removes only issues they agree are unwanted. It checks for duplication, needless abstractions, excessive error handling, and similar readability problems.