auxly

Home dashboard for all your Auxly runs. Scans the working directory for past and present runs (council plans, execute/console runs) and shows them as a grid of cards — title, kind, status, and stages — in a Board tab of the shared Auxly Console. Use when the user wants an overview across runs: "show all my runs", "open the board", "what runs do I have", "the Auxly home/dashboard". Spans many runs (unlike the single-run execute/review views). Standalone; needs only Python + a browser.

Summarize an Auxly run into a clean Markdown digest — what was planned, what shipped (phases/slices done vs total), blockers hit and how they were resolved, warnings raised, check results, review findings, and token/cost — then render it as a Summary tab in the shared Auxly Console and save it to digest.md. Use at the end of a run, or when the user asks to "summarize this run", "what happened", "write a recap/changelog", or "give me the digest". Works on the live console state or a saved run's state.json, standalone or after any Auxly skill.

Execute an accepted implementation plan inside the live, Auxly-branded multi-stage Console — phases, slices, and a progress bar that update in real time as Claude works through the plan, plus an Agents & Models panel showing which agents/subagents are active vs idle, and a Checks panel for build/test/lint gates. Blockers that need a human decision appear as RED notifications requiring action (the user types a resolution in the UI and execution resumes); warnings appear AMBER and dismissable. Renders into the same shared Auxly Console tab as plan/verify/review (no new tabs). Use whenever the user accepts a plan and wants to execute with live monitoring, or says "run the plan with a dashboard", "execute and show progress", or "track execution". Pairs with /auxly-llm-council (feed its final-plan.md) but also works standalone with any final-plan.md or plan.json.

Auxly-branded multi-member CLI planning council. Convenes any mix of installed planner CLIs (Codex, Claude Code, Gemini, agy/Antigravity, OpenCode, or custom) to produce independent implementation plans, anonymizes and randomizes them, then judges and merges them into one final plan — monitored live in an embedded, dark, Auxly-themed web UI that shows which agents and models are on the council, renders readable Markdown (never raw JSON), and surfaces Risks, Pros, and Cons for every plan. If the user has none of codex / gemini / agy installed, it automatically falls back to a Claude-only multi-agent council (architect + pragmatist + risk hawk). Use whenever you need a robust, bias-resistant planning workflow, structured outputs, retries, and graceful failure handling across multiple agents.

Live token (and optional cost) meter for an Auxly run. Records per-agent/per-model token usage and shows a running total — tokens in/out, number of models, and an approximate $ cost — in the header of the shared Auxly Console. Use whenever the user wants to track token usage or spend across a council/execute/review run, asks "how many tokens", "what's this costing", or "show a usage meter". Cost is shown only when an optional prices.json is provided (tokens are always shown); estimates only. Works alongside any Auxly skill or standalone.