xtrm-dev

Bootstrap a complete security pipeline (Dependabot + OSV + Semgrep + gitleaks + pre-commit hooks + Codex review + pr-review-gate) on any GitHub repo. Designed for free user-private repos where GitHub Advanced Security is unavailable. Reusable across…

How to start, sustain, and hand off work in an xtrm-equipped session — orienting on a live board, choosing whether to work solo or delegate, arming continuity so long work survives session death, coordinating panes and specialists, and writing a handoff a…

Cross-stack health verification with runtime-inferred routing. Queries Prometheus + Grafana live via the `mcpq` CLI (sidecar registry lives in `.mcpq.json` — `mcpq servers` lists them) to enumerate firing alerts, down containers, and stale freshness feeds;…

Structured planning skill for xtrm ecosystem projects. Creates a well-documented bd issue board from any task, feature, spec, or idea — with phases, dependencies, rich descriptions, and integrated test coverage via test-planning. MUST activate whenever the…

Help the operator coordinate work across N concurrent tmux sessions (Claude Code, pi, raw shells, vim, REPLs). Inventory state, hand off tasks cleanly, prevent messy-run failure modes, keep hygiene. Not an agent harness; not a /using-specialists replacement;…

Audit, clean, and rewrite CLAUDE.md / AGENTS.md / agent instruction templates so they stay compact, current, and routing-oriented instead of becoming dumped CLI manuals. Use whenever the user asks to update CLAUDE.md or AGENTS.md, clean bloated agent docs,…

Author dynamic multi-agent workflows (the Workflow tool) that inherit specialist semantics without inheriting a fixed chain. Use when work needs fan-out over a discovered work-list, a set-computation barrier, adversarial self-correction, or gates whose…

Portable PR-review discipline usable from any harness (Claude Code, pi, Codex, raw shell). Fetches the PR narrative, the diff, existing review threads, and Codex findings via `gh`; produces a structured issue list under a confidence rubric; and either prints…

Canonical specialist orchestration skill. Use proactively for substantial work that should be delegated, tracked, reviewed, fixed, tested, or merged through specialists: code review, debugging, implementation, planning, doc sync, security checks, multi-step…

Use this skill when creating or fixing a specialist definition. It guides the agent through writing a valid `.specialist.json`, choosing supported models, validating against the schema, and avoiding common specialist authoring mistakes.

Reconcile specialists runtime drift and xtrm-managed asset drift across one repo or many. Use this skill when user says "update specialists", "specialists is out of date", "xtrm drift", "skills not updating", "assets out of date", or when operator needs…

Analyze specialist KPI data in observability SQLite. Use for runtime, payload, waiting, tool-call, and outlier analysis. Token estimates use cl100k_base-style approximation with ~±5% accuracy.

Use this skill for node-coordinator behavior. The coordinator is a CLI-native orchestrator that drives NodeSupervisor via `sp node` commands.

Operator-offline autonomous orchestration overlay. Activate when the user says "auto mode", "full auto", "run autonomously", "I'll leave you alone", or similar — and hands over a multi-item priority list. Layers on top of `using-specialists`: paranoid pacing,…

First-run setup workflow for a Specialists install. Use when the user says "setup specialists", "configure specialists", "change the specialist models", "models shipped do not exist on my machine", "set notes_mode globally", "opt out of GitNexus for one…

Use this skill for synchronous one-shot specialist invocations via `sp script` (CLI) or `sp serve` (HTTP daemon). These run READ_ONLY, template-driven specialists with `$var` substitution and return JSON in-process — no beads, no chains, no worktrees, no job…