AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
badmuriss avatar

badmuriss

14 listings · 0 installs

Open-source publisher. Listings imported from github.com/badmuriss — credited to the original author with their license.

↗ github.com/badmuriss
14 results
Self-run
SKILL

Remove Ai Marks

>

0
18
Free
Self-run
SKILL

Impl

Implement an existing OpenSpec change with executable checks, resumable state, and bounded delegation. Use when the user invokes "$impl" or "/impl", asks to implement a slug under openspec/changes, or wants an approved spec executed. Do not invent a missing spec.

0
19
Free
Self-run
SKILL

Spec

Plan a codebase change with an architecture lens, resolve decisions, and emit an OpenSpec change ready for implementation. Use when the user invokes "$spec" or "/spec", asks to spec or architect a change, or wants proposal.md, design.md, and tasks.md. The optional --council flag adds one bounded review. Do not implement code.

0
13
Free
Self-run
SKILL

Frontend Visual Validation

Validate every rendered frontend change on its actual supported platforms with reproducible browser screenshots and vision review. Use whenever Codex creates, edits, fixes, refactors or reviews UI, CSS, responsive layouts, routes, components, visual states, interactions or frontend assets, and before declaring any frontend implementation complete.

0
14
Free
Self-run
SKILL

Grill With Docs

Grill a plan against the documented domain model, updating CONTEXT.md and ADRs inline. use_when: stress-test a plan against the project's own language and recorded decisions. do_not_use_when: there is no domain documentation, use grill-me.

0
10
Free
Self-run
SKILL

Ingest

Converts PDF, Office, EPUB, image, audio or repo into structured markdown before analysis. use_when: a document is uploaded or referenced, or before summarizing, quoting or analyzing one (\"converte esse PDF\", \"le essa planilha\"). do_not_use_when: the text is already markdown in the repo.

0
18
Free
Self-run
SKILL

Thermo Nuclear Code Quality Review

Perform a strict, read-only maintainability review of a completed implementation diff. Use after impl, before PR preparation, or when asked for a thermonuclear, harsh, deep code-quality, architecture, spaghetti, abstraction, or large-file review. Report evidence-backed findings; never edit code or turn subjective preferences into automatic blockers.

0
12
Free
Self-run
SKILL

Writing

Clear writing standards based on Zinsser. use_when: documentation, commit messages, PR descriptions, technical communication. do_not_use_when: prose for a human reader, use unslop.

0
11
Free
Self-run
SKILL

Rule Curator

Use when auditing, pruning, or cleaning up an agent's entire existing rule set in one bulk pass across every rule source at once (all CLAUDE.md files, all memory entries, AGENTS.md, ADRs, COORDINATION). This is whole-corpus hygiene; NOT for adding, editing, or fixing an individual rule (just edit that file directly).

0
21
Free
Self-run
SKILL

Incredibly Pretty Websites

Build genuinely beautiful, non-AI-slop websites, React-first, research before code. use_when: designing or rebuilding a site or landing page, choosing type, color, layout or motion, fixing AI-looking UI. do_not_use_when: backend, data or copy work.

0
11
Free
Self-run
SKILL

Scrapingdog

Default paid provider for live public web data: scraping, rotating proxies, Google (Search, Maps, News, Trends, Shopping, Images, Flights, Jobs, Scholar), YouTube, TikTok, X, Instagram, LinkedIn, Amazon and marketplaces. use_when: SCRAPINGDOG_API_KEY exists and the task needs live web data, SERP, transcript, review mining, lead enrichment or price monitoring. do_not_use_when: the data is already…

0
18
Free
Self-run
SKILL

Trim Code Comments

Review source comments and remove low-value comments that only narrate visible code while preserving rationale, constraints, safety context, public documentation, and tooling directives. Use when asked to trim code comments or remove dumb, redundant, obvious, excessive, or AI-generated comments. Do not run automatically after implementation.

0
19
Free
Self-run
SKILL

Readme Pass

Make a public repo README concise, scannable, and agent-first. Use when a README is verbose, hard to scan, missing a copyable agent-install prompt, or needs a presentation pass with a banner, badges, and navigation.

0
16
Free
Self-run
SKILL

Research

Research with verifiable grounding, claim-level source adjudication, provider routing, and an auditable markdown finding. Use when researching a topic, fact-checking, comparing sources, publishing a number/value/superlative, or when the user requests research --council. Do not use when the answer already lives in the repository.

0
21
Free
You've reached the end · 14 loaded