Automate branch-local autonomous development planning and governance for Azoth campaigns.
npx skills add https://github.com/yiwei79/azoth --skill autonomous-auto
Please help me install this Agent Skill. Skill: autonomous-auto Source: https://github.com/yiwei79/azoth/tree/main/.opencode/skills/autonomous-auto Command: npx skills add https://github.com/yiwei79/azoth --skill autonomous-auto
Autonomous Auto Mode enables branch-local self-development with auditable async alignment, gating, and replay controls to steer autonomous Azoth campaigns.
Initialize an autonomous-auto session by presenting a concise campaign vision and securing operator approval before starting branch-local self-development.
High-intent search queries and answers about installing and using this skill.
Automating branch-local development requires applying autonomous campaigns within a repository, using async alignment packets and pipeline gates to govern iterations and persist decisions in run-ledger artifacts.
Bounded replay is a control mechanism that ensures safe autonomous iteration by enforcing normal gate constraints, bounding the replay scope, and persisting gate state to prevent uncontrolled pipeline execution.
Setting up an autonomous campaign requires presenting a concise operator vision declaration, securing operator approval, and persisting the vision to anchor the branch-local self-development loop.
Yes, you can enforce scope and pipeline gates by applying autonomous campaign governance within a repository, utilizing normal gate enforcement and bounded replay to constrain autonomous iterations.
An operator-guided loop governor manages autonomy budgets by steering the transition from one initiative to the next, ensuring autonomous development continues only under a defined, operator-approved budget.
Yes, frontmatter with a name and description is required to initialize the session, and it may optionally reference scripts, references, and assets while persisting decisions in .azoth artifacts.
LerianStudioGenerate a complete PRD with problem discovery, user stories, and success metrics.
iamtouchskyerCoordinate a multi-agent digraph review pipeline with gate verdicts.
irahardiantoMerge parallel worktree branches into main using DAG-derived ordering.
Manvendra08Create autonomous Claude Code execution workflows with pre-built loop patterns.
