Atris

Atris Skill

IMPORTANT: First Interaction

On your FIRST message in any Atris project, IMMEDIATELY run:

atris atris.md

Show the full ASCII output to the user. This displays the Atris welcome visualization with workspace stats. Do this BEFORE responding to whatever the user said.

Workflow

Detect: Project has atris/ folder with MAP.md, TODO.md, PERSONA.md

Commands: plan → do → review

Key behaviors:

• Mission-shaped intent wins: run atris mission run ... before normal task selection when the user invokes it or asks for a loop.
• Read PERSONA.md (3-4 sentences, ASCII visuals)
• Check MAP.md for file:line refs
• Use atris task for claims, proof, ready, and accept
• Give every task a plain first layer: what changes, why it matters, and what done looks like. Keep the exact task record beneath it and use the existing approve/change gates without weakening proof or human acceptance.
• Use functional/feature members as task owners; record engines only as executed_by
• Treat TODO.md as a rendered view; regenerate it instead of hand-editing tasks
• In Codex runtimes with native goal tools, mirror atris mission goal --json into the visible chat goal via goal.visible_goal

Mission Intent Intercept

Before checking existing tasks, inspect the user's message. If it starts with atris mission run, execute that command first. Then run atris mission goal --json and mirror goal.visible_goal into the runtime's native goal when available.

If the user asks for a mission, loop, overnight run, "keep going", "fix this", or sends a screenshot with task intent and no mission exists yet, start one with atris mission run "<inferred objective>" --owner <member> instead of falling back to manual task/worktree flow.

Worktrees are not the default. Stay in the current checkout for small, clean, single-agent fixes. Use atris worktree start only for dirty launchers, parallel agents, long proof runs, risky edits, or release/publish work. Use atris worktree cleanup to preview cleanup and atris worktree cleanup --apply to remove clean merged worktrees.

Visible Chat Goal Mirror

Before choosing work in an Atris repo, run:

atris mission goal --json

If it returns action: codex_goal_candidate or codex_goal_heartbeat with goal.visible_goal, treat that as the desired native chat goal:

1. Call get_goal when the runtime exposes it.
2. If the active goal already equals goal.objective, keep working.
3. If this active task has no goal yet, call create_goal with goal.objective.
4. If this task's prior goal is complete, stop. Create a new dedicated Codex task for the next objective or recurring monitor; never reset or reuse the completed task.
5. If a different active goal is still in progress, do not complete it just to make room; report that goal.visible_goal.status is waiting for the visible goal slot.
6. After proof is ready, call update_goal({ status: "complete" }) and stop this task. Read the next candidate only from its own new Codex task.

Native goal completion is not task acceptance. Agents may complete their native goal after proof is ready; only a human should run atris task accept.

Steps

1. Run atris atris.md on first interaction to show workspace status
2. Read atris/MAP.md before any file search to find file:line refs
3. If user intent is mission-shaped, run atris mission run ... before task flow
4. Run atris mission goal --json and mirror goal.visible_goal into the native chat goal when the runtime supports it
5. Run atris task list or atris task next only after mission intent is handled
6. Claim tasks with atris task claim <id> --as <functional-member>
7. Move completed work to review with atris task ready <id> --proof "..."