test-driven-development-tdd
Write the test first, watch it fail, write minimal code to pass
Discover Data & AI skills, compare source repositories and read the instructions for each workflow.
Write the test first, watch it fail, write minimal code to pass
Receive and act on code review feedback with technical rigor, not performative agreement or blind implementation
Check all external inputs for validity - garbage in, nothing out, never garbage out
Maintain skills wiki health - check links, naming, cross-references, and coverage
Write comments explaining WHAT and WHY, never temporal context or history
Lightweight, git-native issue tracking (aka beads) for AI agents. Use for creating, planning, updating, and tracking issues with dependencies. Invoke when user mentions tbd, beads, to-do lists, planning, tracki...
Clarify ambiguous requests by researching first, then asking only judgment calls; stop before implementation.
Create micro-commits (minimal incision) with at least one validation signal per commit.
Generate iterative `plan-N.md` files in the repo root (max N=5). Use only when asked; clarify before `plan-1.md`; stop at `plan-5.md`.
Preflight bead work: verify beads are in use, confirm explicit start/continue, identify active bead or select next `bd ready`, clarify done criteria, and scope audit before marking `in_progress`.
bd (beads) issue-tracking workflow; use when `.beads/` exists or when running `bd` commands.
TRACE (Type/Readability/Atomic/Cognitive/Essential) lens for explaining code and proposing improvements (no edits applied).
Software surgery: contract → invariants → creative search → inevitable incision; minimal diff, compositional core, proof signal.
Review+fix protocol with safety guardrails (unsoundness, invariants, footguns, incidental complexity); requires a validation signal.
Unified shipping + TRACE self-review protocol (beads, proof, PR). Explicit-only.
Convert a project plan written in a Markdown file into a complete bd beads graph (tasks, subtasks, dependencies) using only bd for creation and edits, then review and optimize each bead. Use when asked to beadi...
Algebra-Driven Design (ADD): model domains via minimal algebra + laws; encode in types; test laws.
Gauntlet for absolute claims (always/never/guaranteed/optimal); pressure-test, then refine with explicit boundaries.
Finalize work after validation: confirm a signal, capture proof in PR description and bead comment (if beads), and open a PR (no merge). Use when asked to ship, finish, or open a PR.
Spawn Codex CLI sub-agents non-interactively with fixed model/effort and required flags (web search, skills, yolo). Use when dispatching a single agent per bead via codex exec with a positional prompt string an...
Automate Chrome/Chromium via CDP: navigate, click, fill forms, evaluate JS, screenshot, scrape.
Zig runbook: performance (SIMD + threads), build/test, comptime patterns, allocators, build.zig/build.zig.zon, zero-copy parsing, C interop.
PR autopilot: create/manage PRs, enforce CI gates, apply surgical fixes, and squash-merge via `gh`.
Create a new skill that encodes a team’s bead-implementation method and packages it as a .skill. Use only when the user explicitly says "$method".
The default order uses GitHub repository stars from our indexed data. Stars belong to the source repository and may be shared by multiple skills. They indicate repository interest, not skill installs or an independent quality score.