lastlight

Keep Last Light's docs in sync with the code. Use before committing changes to apps/server/workflows/, skills/, config/default.yaml, src/connectors, src/state, src/engine/router.ts, src/config/, packages/cli/src, packages/shared/src (providers/overlay helpers), or agent-context/ — or whenever the docs-check pre-commit hook fires. Maps each changed file to the doc surfaces it affects (the in-repo apps/server/spec/*.md AND the apps/www site) and updates them.

Install the `lastlight` CLI and connect it as a CLIENT to an existing Last Light server — log in, save the token, and verify the connection. Use when the user wants to "connect / point my lastlight CLI at a server", "log in to Last Light", "set up the lastlight client", or run lastlight commands against a remote instance. For standing up the server itself use lastlight-server; for editing a deployment's config use lastlight-overlay.

Debug a running Last Light instance over its admin API with the `lastlight` CLI — no SSH. Use when the user wants to "look at / debug a failed or stuck workflow", "why didn't the bot do X", "check a bad agent run / phase error on <instance>", "tail a session", "search the logs", "read the server/docker logs", "list/trigger a cron", "resolve an approval", or "check stats/cost" against a deployed server. Needs a CLI already connected (see lastlight-client to log in). For standing up the server use lastlight-server; for editing config/workflows use lastlight-overlay.

Orientation & router for the Last Light skills. Use ONLY when the user has NOT named a concrete Last Light task — i.e. they're unsure which skill/flow they need, ask "what can Last Light do / where do I start", say "help me with Last Light" with no specific goal, or want a guided tour across server / client / overlay / debug / evals. Do NOT use when the ask already names a task — "set up/deploy a server" → lastlight-server, "connect/log in my CLI" → lastlight-client, "customize a workflow/prompt/persona/config" → lastlight-overlay, "debug a failed/stuck workflow or run on an instance" → lastlight-debug, "run evals / compare models / author a case / build a PR-review dataset" → lastlight-evals, "iteratively improve an eval toward a target score" → lastlight-evals-loop: invoke that skill directly. This skill only routes and, when the goal is ambiguous, asks. Also invocable as /lastlight-guide.

Install and configure a Last Light SERVER — the GitHub maintenance agent plus its docker-compose stack — on a host. Use when the user wants to "set up / install / deploy / stand up a Last Light server or instance", configure its GitHub App, models, managed repos, or domain, or get the agent running for the first time. Drives the `lastlight` CLI; for connecting an existing CLI to a server use lastlight-client instead, and for editing a running deployment's config use lastlight-overlay.

Drive a Last Light EVAL toward a target score with a disciplined, anti-gaming improvement loop — run → mine failures → propose candidate fix(es) → re-measure → keep the best or revert → repeat. Use when the user wants to "improve / raise the pr-review F1", "make the reviewer better against the eval", "close the loop on evals", "iterate on prompts/skills to pass more cases", or "tune the workflow to hit a score target". The loop diagnoses on a TRAIN split and validates on a BLIND held-out split so fixes must generalize, not overfit; it prefers generic overlay prompt/skill edits, and stops for human sign-off before editing any gold answer. For a one-off run/compare use lastlight-evals; to hand-fork a workflow/prompt use lastlight-overlay. Needs an already-scaffolded evals workspace (lastlight-evals) with a pr-review dataset.

Scaffold, configure and run a Last Light EVALS workspace — the harness that runs Last Light's real workflows against a mocked GitHub and grades them deterministically. Use when the user wants to "set up / scaffold Last Light Evals", "create an evals workspace or instance", "run evals", "compare models", or author new eval cases — including "create an eval dataset/case from this PR/issue <github url>" and "create a PR-review eval dataset from these gold PRs" (triage / code-fix / pr-review instances, via `add-case --pr <url> --review|--code-fix` / `--issue`). GitHub is mocked when running, so no real GitHub token is needed — only a model provider API key (the one exception is `add-case`, which reads real PRs/issues via `gh`).

Create or customize a Last Light deployment OVERLAY (the private instance/ repo) — scaffold it, then fork built-in workflows, prompts, skills, or the agent persona (agent-context) so a deployment can override them. Use when the user wants to "create a Last Light overlay / instance repo", "customize / fork a workflow", "override a prompt or skill", "change the agent's persona/soul/rules", or tune a deployment's config without forking the whole codebase. For first-time server install use lastlight-server.

Drive a real headless browser against a locally-served web UI and capture screenshot evidence, reporting step-level PASS/FAIL/BLOCKED with extracted DOM text and console errors. Use when a phase on the docker QA image must exercise a rendered UI and attach screenshots — it complements the text-evidence verify/qa-test skills.

Install dependencies and run the test/lint/typecheck gate for a change inside the sandbox — package-manager detection, install-first, and (when implementing) TDD discipline. Use when implementing, fixing, or verifying code in a pre-cloned repo.

Conversational assistant for messaging-platform threads (Slack, Discord). Answer questions about repos, PRs, and issues, explain code, and guide users to natural-language triggers (e.g. "build owner/repo#N", "triage owner/repo", "review PRs on owner/repo", "status").

The shared rubric for reviewing a code change — precision-first, high-signal findings only (Critical/Important), plus what to check (correctness, security, edge cases, regression risk, test coverage). Use when reviewing a PR or a branch diff.

Debug a running Last Light instance (usually production) via the `lastlight` CLI instead of SSH. Use when investigating a failed/stuck workflow, a bad agent run, a phase error, or "why didn't the bot do X" — anything that previously meant SSHing in to read SQLite or session logs.

Record a short demo VIDEO of a PR or feature — drive the repo's web UI in a real headless browser, capture the session, and composite a titled, size-capped mp4 with ffmpeg. Use on the docker QA image when the deliverable is a playable demo clip (single walkthrough or before/after comparison), not a text/screenshot report.

Answer a question directly — a sourced, neutral reply to an information/explanation/comparison request, from a GitHub issue or a Slack thread. Research repo docs and the web, output the answer, label `question` (GitHub only), leave open. Never write an agent brief, mark ready-for-agent, or change code.

Handle a non-build maintainer comment on an issue or PR — close, reopen, label, dedupe, answer a brief question, or triage. Action-only; redirect anything that needs code changes to /build.

Triage GitHub issues through the canonical state machine — classify, label, deduplicate, request missing info, write agent briefs, and manage stale issues. Use on demand or on a triage cron.

Answer a maintainer's question about an open PR with concrete, code-cited evidence. The PR-side counterpart to issue-comment — for questions tied to the diff, not a full review.

Review a GitHub pull request and post one formal review — advance the existing discussion and give precision-first, high-signal feedback. A pure code review — no building. Use when asked to review a PR or on a cron PR scan.

Run an automated QA flow against a CLI or a locally-served app and report step-level pass/fail with evidence. Use when asked to QA-test a feature, exercise a flow end-to-end, or smoke-test what a PR changed.

Generate a health report for a GitHub repository — open-issue and PR backlog, unreviewed PRs, stale needs-info, failing CI, and the resulting action items. Use for a status report or on a weekly cron.

Process a maintainer's comment on a security scan-summary issue — break selected findings out into individual actionable issues, or record accepted risks / false positives in SECURITY.md.

Diff-scoped security review of SDLC concerns GitHub's scanners miss — workflow/CI hardening, auth changes, secret handling, supply-chain churn. Files one dated summary issue with a task-list of findings. Use on a security cron or when asked to scan a repo.

Test a behaviour claim as an investigator and report whether the evidence confirms or refutes it — CONFIRMED / REFUTED / INCONCLUSIVE with bash-captured evidence. Use when asked to verify a claim, prove a fix works, or check that a PR does what it says.