hmbseaotter
UserWrite software specifications an AI coding agent can execute: EARS/SHALL requirements, a determinism boundary, and a /specify interviewer that emits the spec + build prompt (or the skill/agent artifact).
Categories
Indexed Skills (3)
specify
Interview the user to produce a complete, agent-ready specification, then emit what is needed to build it: for a coded feature/agent a phase-scoped build prompt, or for a skill / declarative agent the artifact itself (SKILL.md / AGENT.md). Surfaces every assumption for review, slices the work into a phased plan, and runs a build-readiness (model/effort) check. Use when the user wants to design, specify, plan, or scope a new agent, feature, or skill; turn a vague idea, PRD, or plan into a rigorous spec; write requirements in EARS / SHALL with testable acceptance criteria; produce a build prompt for a coding agent; or amend, advance, review, audit or sweep an existing spec under specs/ for drift and staleness. Reasoning-driven, one question at a time.
csv-column-summarise
Profile the columns of a local CSV — inferred type, fill rate, cardinality, numeric min/max/mean, and sample values, plus a one-line plain-English guess of what each column means. Use when the user points at a .csv and wants to understand its columns without opening a spreadsheet.
meta-prompt
Use when the user wants help writing, generating, or improving a prompt for an LLM - e.g. "write me a prompt for...", "turn this into a prompt", "help me prompt for...", "improve this prompt", or when they hand over a use case and want a polished, reusable prompt back. Produces a finished, copy-ready prompt (not advice about prompting). Also applies when the user references the CoSTAR framework or asks for a prompt built to a named structure.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.