tobywise
- 14 skills
- 0 followers
- 2 hours ago last updated
- ▌ Debrief · tobywiseProduce an evidence-based LangSmith debrief or incident postmortem for a Paseo agent, thread, or repository. Use when asked to debrief a run, diagnose what happened in an agent session, or write a postmortem.
- ▌ Planning · tobywise bundleThe shared format and rules for implementation plans. Use when writing, splitting, revising, or interpreting an implementation plan, including when implementing or reviewing work against one.
- ▌ Prototype · tobywiseBuild and optionally run one rapid, exploratory, single-use scientific analysis prototype, without production engineering ceremony. Use when the user wants a quick exploratory analysis, a throwaway scientific prototype, or invokes /prototype.
- ▌ Run Checks · tobywise bundleHow to run tests, builds, type checks, installs, and other verification commands on this machine — locally, or on remote Crabbox workers when they are available. Use before running any test suite, build, type check, dependency install, migration, Docker build, or other expensive command. Also says which tests to run while iterating and when the full suite must run.
- ▌ Slide Deck · tobywiseCreate a navigable HTML slide deck to illustrate problems or concepts. Use when the user wants a visual aid to explain a complex topic, or to illustrate a problem or design.
- ▌ Orchestrate · tobywiseA flexible guide for carrying out a plan from the main session by delegating implementation and review to agents and running verification yourself. Use when asked to implement and review a plan, carry a plan through to completion, or when the user invokes /orchestrate.
- ▌ Plan Review · tobywise bundlePresent a proposed plan with several real decisions in a local HTML page, collect answers, and revise it in rounds before writing the final plan. Use when the user wants to decide options or review a plan visually.
- ▌ Pr Description · tobywiseDraft a reviewer-focused pull request description from the current branch changes. Explain what changed and why in short, scannable prose, classify scientific impact as NONE, MINOR, or MAJOR, identify risks and validation, and choose a compact call diagram, diff-style comparison, pseudocode explanation, or DAG when useful. Use when asked to write, improve, update, or review a PR description or explain a branch diff.
- ▌ Github Pr Review · tobywiseFormat and publish completed review findings as GitHub PR comments or submitted reviews. Use only when the user requests posting to GitHub. For assessing a PR, including requests to review and post, use review-implementation first. Do not select this skill for reviews or drafts requested only in chat.
- ▌ Langsmith Review · tobywise bundleReview LangSmith traces, runs, threads, and OpenCode or Paseo sessions through the read-only helper. Use when asked to inspect, diagnose, summarize, compare, or query LangSmith activity, latency, token usage, tool calls, or errors.
- ▌ Scientific Analysis · tobywiseUse ONLY when planning, implementing, reviewing, or orchestrating a plan whose review standard is SCIENTIFIC_ANALYSIS.
- ▌ Review Implementation · tobywise bundleReview pull requests, local implementation changes, or fixes against the applicable requirements. Use for code reviews and PR reviews, including requests to review and post to GitHub. Report findings with severity, likelihood, and scientific tiers when applicable; use github-pr-review only for the subsequent GitHub submission.
- ▌ Scientific Risk Review · tobywiseIndependently rate the severity and likelihood of proposed blockers or possible scientific risks, judge whether they should block, and whether a blocker is acceptable. Use after a review raises blockers or risk candidates, or when asked for a second opinion on a finding.
- ▌ Create Scientific Rework Plan · tobywiseAudit an existing scientific codebase and plan a safe single-use rework. Use when asked to simplify, rework, or right-size an existing scientific analysis codebase, or when the user invokes /create-scientific-rework-plan.