AI Resources
A small, opinionated toolkit I reach for every day, built for Claude Code — Anthropic's terminal-first agent. Download, fork, adapt; everything here survived production, not toy examples.
01 · Prompts
A prompt is a specific instruction you hand an LLM to produce a specific kind of output. Good prompts are opinionated and constrained: they give the model a role, a format, and an escape hatch when uncertain. Think of each one as a short contract — you know what you asked for, the model knows what it owes you. The prompts in this library are downloadable .md files with YAML frontmatter: drop them into a chat, or copy the body into your own workflow. Nothing is locked to Claude — same contracts work with any capable model.
Ritual for starting a new Claude Code session. 5 minutes of reading before touching anything — in a set order. Solves the "cold start bias": the agent jumps to code before understanding state. List of sources of truth, reading order, infra checklist.
Three paths for making decisions: (1) one sensible option → proposal + "see any gaps?", (2) 2-3 real options → classic 3-line format + TL;DR, (3) > 3 options → narrow first. Eliminates fake A/B/C and option multiplication beyond need, while preserving discipline for real decisions.
The core working-mode manifest for Claude Code on long-form projects. Normalized version of the personal harness from the GroqeSTT project — auto-loaded via CLAUDE.md on every session. Contains agent-flaw diagnostics, seven non-negotiable rules, pre/during/post-task checklists, red flags, user steers, and the agent's decision-making space protocol (three paths).
Battle-tested prompt for secure OpenClaw installation on an isolated Linux server. Includes threat model based on audits from Microsoft, Kaspersky, Cisco Talos and CrowdStrike, config hardening, sandboxing, and a complete step-by-step plan. Mind the freshness — current as of creation (Q1 2026), CVE versions may change.
Build → Critique → Refine loop for when "good enough" isn't good enough. Instead of hoping the first output is perfect, plan for 2-3 iterations from the start. Each round has a specific focus: first pass for structure and correctness, second for quality and edge cases, third for polish and performance. Works for code, writing, architecture, and design.
Write the spec before you write the code. This prompt enforces a spec → plan → build → verify loop that prevents scope creep, architectural drift, and "it works but not how we wanted" outcomes. The spec is the contract between what you want and what gets built. Every implementation decision traces back to a spec requirement.
Strategy for managing long AI sessions without losing context or quality. Covers when to compact, how to structure checkpoints, what to keep in context vs. offload to files, and how to hand off between sessions. Prevents the "it forgot what we were doing" problem and the "last 20% of context produces worse output" degradation.
Systematic method for understanding an unfamiliar codebase fast. Instead of reading code top-to-bottom, follow the data flow: entry points → routing → business logic → data layer → external integrations. Produces an architecture map, key abstractions list, and "where to look" guide in under 30 minutes. Works for any stack.
02 · Skills
Claude Code is Anthropic's terminal-first AI developer agent. It reads your repo, runs commands, edits files, and keeps a single conversation across days of work. A skill is how you teach it to behave consistently. A skill is a single Markdown file — SKILL.md — with YAML frontmatter. Claude Code scans the frontmatter when it needs help; if a trigger matches, the body is loaded into context for that turn. You get a playbook the agent follows silently: rules, anti-patterns, when to use, when not to. No config UI. No restart. Just a file.
Kreator i optymalizator CV maksymalizujący szanse na rozmowę rekrutacyjną. Przerabia istniejące CV pod konkretną rolę lub buduje od zera. Zawiera: Extraction Interview (wydobywanie ukrytego złota z kandydatów), Nothing-Left-Behind Principle, Page Density Algorithm, DOCX formatting specs (Classic + Modern), role-specific keyword patterns (Python/ML/QA/DevOps/Frontend), Strategic Truth-Stretching Framework, pełną bazę wiedzy ATS/Polish market. Obsługuje wszystkie branże i poziomy (junior/mid/senior), ze szczególną optymalizacją dla polskiego rynku IT.
Skill enforcing Read before every Edit/Write in the same session. Eliminates "blind edits" based on memory — every file the agent modifies must have been read in the current session. Protects against overwriting others' work.
Skill enforcing plan-first: a plan presented to the user before any code change. You wait for GO, then write. Exception: batch approval ("do X and commit"). Eliminates "I already started" and unauthorized commits. Two formats: single-path proposal (default for most steps) and plan-with-variants (when there are real options).
Skill for generating the next-session-phase{N+1}-brief.md file before ending a session. Eliminates the gap between sessions — the next session has ready context, first task, and infra quick-check. Zero momentum loss from "where did we leave off".
Meta-analytical meeting breakdown — not a summary, but an efficiency analysis. Produces signal/noise ratio, decision rate, topic weight distribution, voice distribution, red flags, and a brutal one-sentence Shelby-Musk verdict. Use for meetings over 10 minutes when you need to understand where the time actually went.
Framework for running retrospectives that actually change something. Produces a facilitation plan with warm-up, data gathering, insight generation, and action items. Includes 10+ retro formats to prevent format fatigue. The output is not a feel-good session — it's a systematic process for turning team friction into concrete improvements with owners and deadlines.
Framework for running effective 1:1 meetings as a manager, mentor, or peer. Produces an agenda with the right questions for the context, a note-taking template, and follow-up actions. Covers career development, performance feedback, blockers, relationship building, and difficult conversations. The output is not corporate theater — it's a tool for building trust and getting real signal.
Audience-calibrated messaging for different stakeholders. Transforms the same information into messages tailored for executives (business impact), engineers (technical detail), clients (value delivered), or team (actionable next steps). Produces ready-to-send drafts with tone, length, and emphasis matched to the recipient's priorities and communication style.