Roger-luo

Use when hardening Ion's user experience by driving the ion CLI end-to-end against a real project and improving what feels rough — i.e. "dogfood ion on <project>", "use ion in X and fix the UX gaps", "is ion actually pleasant to use", "does ion guide the agent to the next step", or any shakedown of the CLI from a user's seat. Dogfooding judges three things at once: (1) human terminal UX — is plain (non-JSON) output friendly and readable; (2) progressive prompting — does every command's output (human text AND the --json envelope) make the next step obvious so a user or agent never stalls; (3) workflow ergonomics — does a real goal take too many commands, calling for a new or consolidated command rather than another hint. Covers driving the full skill lifecycle on a scratch repo and an established one, logging each rough edge, and FIXING the CLI (next-step hints, richer action_required envelopes, friendlier output, new/merged commands) with a failing-test-first discipline using the scenario crate + integration

{{ description }}

{{ alpha_description }}

{{ beta_description }}

{{ description }}

{{ description }}

{{ description }}

A test skill

Create a project-specific RFC-writing skill named `new-rfc` by interviewing for RFC conventions, applying a reusable template, and validating the generated skill.

Draft and revise RFC markdown files for agx using project template, metadata conventions, and review gates.

Run a structured requirements interview to remove ambiguity before implementation. Use when users ask to be interviewed, ask for guided questions, mention ask-user-question or interview skills, or when key decisions are missing and proceeding would risk rework.

Use when running autotune — the autonomous, metric-driven code-tuning CLI (assumed already installed) — on a project to improve a measurable metric such as benchmark speed, test coverage, binary size, or model accuracy (e.g. "use autotune to speed up X", "tune this project with autotune", "have autotune improve my benchmark/coverage"). Also use when an autotune run misbehaves (crashes, hangs, banks no real improvement, scores real wins as noise) or lacks a config option / adaptor / behavior your project needs.

Use when hardening the autotune CLI by running it end-to-end against a real target repo — i.e. "try autotune on <project> and fix what breaks", "run autotune against X and watch for malfunctions", "can autotune actually improve X", or any dogfooding/shakedown of the tune loop. Dogfooding has two goals: (1) robustness — the loop doesn't crash, hang, orphan processes, fill the disk, or record the wrong outcome; and (2) efficacy — autotune produces meaningful, correct improvements (good hypotheses, real wins that clear the noise envelope, not gamed/overfit). Covers launching a traced unattended run, watching the trace/logs/ledger, diagnosing via trace `phase.decision`, fixing bugs OR improving autotune itself (research prompts, scoring, specialization) with a failing-test-first discipline, and rerunning until N iterations complete cleanly and the wins are judged real. Not for plain unit changes — just run nextest.

Project management CLI for tracking initiatives, projects, and tasks across GitHub repositories as a recursive directory hierarchy backed by a local git repo, with bidirectional GitHub issue sync and LLM-powered triage. Use this skill whenever working inside an armitage org directory (has armitage.toml at root), managing roadmap nodes, triaging GitHub issues, syncing with GitHub, or planning project structure. Also use when the user mentions armitage commands, node.toml files, triage workflows, or issue classification.

A CLI tool for creating, managing, and validating RFCs (Request for Comments) in a project. Invoke with `ion run rfc-writer`.

Create, edit, and structure terminal-based slide presentations using presenterm markdown format. Covers slide syntax, commands, code highlighting, columns, images, themes, and speaker notes.