Updated 6d ago
by eliasstravik
skills
Nine eval-built agent workflows for planning, implementation, copy, and repository work.
21 GitHub stars (21)
Steady at 21 stars — no new stars in the 16 days since Sep 30.
- 12 skills (12 skills found in its SKILL.md files)
- 0 forks (0 forks)
- 6d ago last update (Last updated 6d ago)
Install
Skills CLI
npx skills add eliasstravik/skills
Asks which skills and agents to install. Just one: npx skills add eliasstravik/skills --skill autoconsultant — add -g for every project.
Claude Code
git clone https://github.com/eliasstravik/skillscp -r "skills/skills/autoconsultant" ~/.claude/skills/
Copies the autoconsultant skill into ~/.claude/skills/ for every project (use .claude/skills/ for one project). Repeat for the other skills listed below.
Plugin
/plugin marketplace add eliasstravik/skills
Run inside Claude Code if the repository ships a plugin marketplace, then pick plugins from /plugin.
Skills can run code on your machine — read the SKILL.md first.
What it does
In the author's own words, from each skill's SKILL.md.
-
autoconsultant
Autoconsultant owns the approval-gated workflow that takes an ambiguous repository change through a persistent Consultant consultation and adversarial plan review, then prepares separately approved, independent Optimizer and Evaluator instructions plus one copy-paste-safe handoff for a fresh agent. It triggers when a request asks to combine consultation or plan approval with Automake or an evalua…
skills/autoconsultant/SKILL.md on GitHub (opens in a new tab) -
automake
Automake owns a Git-backed ratchet for safely improving a repository through separate optimizer and evaluator roles: it establishes a baseline, proposes and checks candidates, obtains an independent comparison against the current committed state, keeps accepted changes, and restores rejected ones exactly. It triggers when the request is to prepare or execute this bounded improvement workflow, dra…
skills/automake/SKILL.md on GitHub (opens in a new tab) -
consultant
Use for persistent, one-question-at-a-time consultation on an ambiguous repository change: inspect supplied territory, save decisions, produce an approved plan, run a scored bounded adversarial review with explicit targets and automatic-review limits, resolve or accept human-owned risks, and hand off to a fresh builder without implementing. Trigger on explicit /consultant or requests for durable…
skills/consultant/SKILL.md on GitHub (opens in a new tab) -
copywriting
Triggers when a request asks for customer- or audience-facing commercial copy for a product, service, or organization, including homepage, landing-page, sales, advertising, launch, social, newsletter, or marketing email copy; excludes internal, personal, editorial, research, strategy, planning, journalist-pitch, and implementation requests.
skills/copywriting/SKILL.md on GitHub (opens in a new tab) -
demo
The demo skill owns turning a product concept, release artifact, checklist, or named demo into a convincing, self-contained local try-it experience. It stages only missing product context with deterministic, visibly simulated behavior; builds or revises the focal interaction; serves it on loopback; and verifies the exact preview, restart path, and QA result. It triggers when a user asks to build,…
skills/demo/SKILL.md on GitHub (opens in a new tab) -
grilling-fix
Triggers alongside the grilling skill whenever the user invokes grilling-fix (e.g. `/grilling /grilling-fix`) or asks to be grilled with lettered options. Overrides only grilling's answer-option format; every question must offer explicit options labeled with letters (A, B, C…), never numbers, plus grilling's usual recommendation, and the user may always answer in their own words. Not for use with…
skills/grilling-fix/SKILL.md on GitHub (opens in a new tab) -
html
The html skill owns creating and revising durable, standalone HTML artifacts: a complete self-contained index.html saved in a requested or generated artifact directory, with inline dependencies, accessible structure, loopback serving, and exact-file verification. It triggers when a user asks to create, build, turn content or data into, save, revise, or update a “durable HTML artifact,” “one-file…
skills/html/SKILL.md on GitHub (opens in a new tab) -
midwit
The Midwit skill is the user’s ongoing anti-complexity gate for a conversation. It triggers on /midwit or requests to maintain an anti-overengineering mode, keep the user honest across tasks, challenge elaborate infrastructure before work, or propose doing nothing when the goal is already met. It asks for a ruling only when it has a specific simpler or null path, quantifies that path as 10–100x l…
skills/midwit/SKILL.md on GitHub (opens in a new tab) -
prototyping
Triggers when a user asks to prototype a decision, compare visible or behavioral options in throwaway pages, keep prototyping through a consultant, grilling, or wayfinder session, or says /prototyping. It builds one loopback session that grows step by step and also handles a one-off "prototype this" request. Not for durable standalone HTML artifacts, polished product demos, browser QA, or integra…
skills/prototyping/SKILL.md on GitHub (opens in a new tab) -
skill-issue
Triggers when an agent authors, reviews, names, or decides whether to create an agent skill, including requests to write SKILL.md, choose a skill type, evaluate skill-worthy content, or split material into references/scripts. Not for running skill evaluations or description optimization, which belong to skill-creator.
skills/skill-issue/SKILL.md on GitHub (opens in a new tab) -
voice-mode
The voice-mode skill owns the conversation’s delivery and interaction behavior when the user needs to listen instead of look at a display. It triggers on explicit “/voice-mode” or “activate Voice Mode,” and on concrete hands-free context such as “I’m driving,” “I’m away from the screen,” “I’m wearing headphones,” or “keep every reply spoken,” especially when the user requests concise conclusions,…
skills/voice-mode/SKILL.md on GitHub (opens in a new tab) -
queue-triage
Classifies fictional support requests for a sample operations team.
evals/demo/fixtures/agent-skill/queue-triage/SKILL.md on GitHub (opens in a new tab)
No skill matches that filter.
Review before you install
A skill is a set of instructions and scripts your agent will follow, and it can run code on your machine. Read the SKILL.md and any scripts first — this listing is not a security review.