Log moderate-confidence decisions to DEVLOG during autonomous software project runs.
npx skills add https://github.com/berad217/human-training --skill leroy-jenkins
Please help me install this Agent Skill. Skill: leroy-jenkins Source: https://github.com/berad217/human-training/tree/main/skills-source/leroy-jenkins Command: npx skills add https://github.com/berad217/human-training --skill leroy-jenkins
This Skill eliminates wasted AI token quota near 5-hour or weekly reset windows, and removes the friction of pausing for confirmation on moderate-confidence decisions during software project work.
Invoke the leroy-jenkins skill with your desired project work goal, or leave the goal blank to select a pre-built canned mode for autonomous project tasks.
High-intent search queries and answers about installing and using this skill.
Use an autonomy-biased operating overlay to convert unused token quota into productive work by running comprehensive project checkups or fast robustness audits on your software projects unattended.
Run autonomous code changes by using an operating mode that logs moderate-confidence decisions to a DEVLOG instead of pausing for confirmation, executing tasks on isolated git branches for safe review and rollback.
The best way to automatically sweep for robustness bugs is to invoke a canned audit mode that uses parallel subagents to perform a fast robustness audit across your codebase.
Yes, you can run unattended long-haul tasks for hours by using subagent delegation to stay context-lean, leaving decision breadcrumbs and isolated git branches for later review.
Review autonomous AI decisions by checking the DEVLOG, which records all moderate-confidence decisions and breadcrumbs made during unattended long-haul runs for easy tracing.
You should avoid autonomy-biased operating modes for high-risk production changes, as it is designed for hobby software development workflows and skips pausing for moderate-confidence decisions.
Shade666Create isolated Git worktrees with priority-based directory selection and safety checks.
zenobi-usVerify tests and merge, push, keep, or discard a development branch.
johnlindquistManage long-running agent sessions with progress tracking and context persistence.
breakingcircuits1337Create AI chat interfaces with agent integration, human oversight, and client-side tools.
