JakeSelby
- 15 skills
- 0 followers
- 6 hours ago last updated
- ▌ Architecture Viewer · jakeselbyOpen or update an architecture diagram with the selected builtin or custom viewer, and preserve a neutral session handoff. Use when the user asks to inspect architecture interactively or continue an existing diagram session.
- ▌ Upstream Contribution · jakeselbyContribute to a repository you do not own — an open-source project, another team's codebase, a vendor's SDK — from a fork, without burning maintainer trust. Use when the working repo has an `upstream` remote, when the user says "open a PR against <someone else's repo>", or before the first commit in any repo the user is a guest in.
- ▌ Code Quality Instruments · jakeselbyMeasure whether a test suite is actually any good, rather than only that it exists and passes. Use when adding or reviewing tests on a change that matters, when a suite passes but a bug still shipped, when coverage is high and confidence is not, and before promising a module is well tested. Carries branch coverage, mutation score, the complexity-times-coverage risk signal, duplication, and the per-language instruments that produce each.
- ▌
- ▌ Design Loop · jakeselby bundleRaise the visual quality of something that already renders — lock a visual target, then loop build → screenshot → independent scored critique → fix, against rubrics with hard accessibility, design-token, runtime and asset-licensing gates. Use when asked to make a UI, page, HTML doc, dashboard, game scene or 3D asset look dramatically better, more polished or more professional; to improve the visual design, look, styling or visual quality of a surface; or to iterate toward a reference, mockup or screenshot. Also use proactively after building or substantially changing a user-facing surface, before calling the visual work done. Not for deciding what screens exist or how flows behave — that is UX specification, handled elsewhere.
- ▌ Plan Authoring · jakeselby bundleWrite or revise a plan file, proposal, design doc or handoff that the user reviews before approving work. Carries the Review Card contract — the one-screen summary, the system diagram, the numbered steps, the decision block — and the addendum rules for everything below it. Use before writing anything into .agent-harness/plans/, and for any markdown deliverable whose job is to get a go/no-go.
- ▌ Spike Contract · jakeselbyFrame a technical spike so its result is a decision rather than an impression — the question, the cheapest experiment, a numeric exit criterion, the measured result, the machine it was measured on. Use when asked to "spike", "prototype to find out", "de-risk", or "check whether X is feasible", and when writing the spikes section of a plan or README.
- ▌ Workflow Status · jakeselby bundleShow progress of background Workflow runs — which agents have returned, which are still working, and how much output has accumulated. Use when the user asks about workflow progress, says "/workflows doesn't work", asks "is the workflow done", "how's the workflow going", "check the workflow", or wants to inspect a multi-agent orchestration run. Reads local Claude workflow journals; Codex background-agent status is not implemented by this reader.
- ▌ API Verification · jakeselbyProve a search, filter or API answer is real before you rely on it, and budget web search across a fan-out. Use when querying an unfamiliar API, when a filter returns suspiciously clean results, when a per-item error might have been swallowed, and when briefing research agents.
- ▌ Licensing Review · jakeselbyVerify a third-party library, asset, model, font, dataset or copied snippet is safe to ship under the project's licensing stance, and record it correctly. Use before adding or upgrading any dependency, before downloading any asset, and before a release.
- ▌ Migration Safety · jakeselbyWrite, review or apply a database schema migration without destroying data. Use for any task that creates or modifies a migration file (Alembic, Prisma, Django, Rails, Flyway, raw SQL), and before applying one to a shared environment.
- ▌ Harness Authoring · jakeselbyDecide where an instruction belongs and write it there. Use when asked to add, change or remove a rule, skill, instruction, hook, setting or CLAUDE.md line, when asked to "remember" something that should persist beyond this session, or when a correction should apply to future sessions. Routes the change into the agent-harness checkout, a personal file, a repo's own instructions, or auto memory, and runs the sync and lint.
- ▌ Delegation Tiering · jakeselbyDecide whether to spawn a subagent and which model tier and reasoning effort it should run on. Use when planning a fan-out, choosing a subagent model, writing a workflow script's opts.model, authoring an agent definition, setting a repo's cost posture, or when a delegation decision is non-obvious. Carries the evidence, the bands, the safety conditions and the untrusted-content protocol behind the standing rule.
- ▌ Transcript Hygiene · jakeselbyBound what subagents return and what tool output enters the transcript; load when briefing a subagent or reading large output.
- ▌ Worktree Per Agent · jakeselbyIsolate an agent's work in its own git worktree branched off the default branch, so two agents never land conflicting changes on the shared checkout. Use at the start of any implementation task in a repo where other agents or the user may also be working, and whenever a repo's instructions say "work in a worktree".