LifeOS

Full implementation lifecycle with adversarial review using subagents

Address PR review feedback and verify independently

Draft and create a GitHub issue from a description, investigation, or conversation context. Use when a problem is too large for a quick fix, when you want to propose a change for later, or when the user asks to "file an issue", "create an issue", "draft an issue", "open a ticket", or "log this for later". Also useful when another skill (like tune) determines a change is too large and needs to be tracked as an issue instead.

Merge a PR and update upstream GitHub issues with progress

Validate a PR against PR standards before requesting review

Health-check the LifeOS persona layer (files load, /chat matches the Telegram bots per persona, no personal values leak into committed persona files, frontmatter valid) and open every persona doc in Sublime for editing.

On-demand routing health check for /chat. Verifies every model (auto / Sonnet / Opus / Gemma-local / Claude Code) on BOTH text and voice, across every persona, and that messaging a persona in /chat is identical to messaging its Telegram bot.

Address PR review feedback and verify independently

Synthesize recent PR activity into a cross-PR net-effects briefing

Draft and create a GitHub issue from a description, investigation, or conversation context. Use when a problem is too large for a quick fix, when you want to propose a change for later, or when the user asks to "file an issue", "create an issue", "draft an issue", "open a ticket", or "log this for later". Also useful when another skill (like tune) determines a change is too large and needs to be tracked as an issue instead.

Full Codex implementation lifecycle for GitHub issues, pull requests, or freeform coding tasks: understand scope, plan, create a branch, write tests, implement changes, run verification, create a PR, perform adversarial review/address rounds, merge, and update linked issues. Use when the user invokes $implement, asks to implement an issue such as '#123', says 'implement this', asks for end-to-end delivery from task to PR/merge, or wants the former Claude /implement workflow in Codex.

Merge a PR and update upstream GitHub issues with progress

Analyze an open-source repo and surface ideas/patterns LifeOS could learn from

Validate a PR against PR standards before requesting review

Run adversarial PR review with specialist subagents

Housekeeping report — find stale PRs, orphan branches, stale issues, and stale plans

Personal daily summary — what you shipped, what's in progress, what needs attention

Tune LifeOS orchestrator behavior based on feedback about a bad response. Use when the user describes a problem with how LifeOS answered a query — wrong sources, too verbose, not persistent enough, poor tool selection, bad source integration, missing detail, gave up too early, etc. Also covers Codex task execution issues (scope, persistence, notification style). Makes a surgical prompt edit, branches, commits, and pushes. Trigger on: "tune", "response-tuning", "fix the prompt", "tune the response", "it didn't search enough", "too verbose", "improve the orchestrator", "the response was bad", or any feedback about LifeOS chat/orchestrator quality.

Comprehensive sync + vault + LLM + service health checkup. Surfaces what's stale, what's broken, what's about to break.

Tune LifeOS orchestrator behavior based on feedback about a bad response. Use when the user describes a problem with how LifeOS answered a query — wrong sources, too verbose, not persistent enough, poor tool selection, bad source integration, missing detail, gave up too early, etc. Also covers Claude Code task execution issues (scope, persistence, notification style). Makes a surgical prompt edit, branches, commits, and pushes. Trigger on: "tune", "response-tuning", "fix the prompt", "tune the response", "it didn't search enough", "too verbose", "improve the orchestrator", "the response was bad", or any feedback about LifeOS chat/orchestrator quality.

Analyze an open-source repo and surface ideas/patterns LifeOS could learn from

Run adversarial PR review with specialist subagents