dhanesh

Extracts a reusable code pattern from the current repo and authors a portable, validated Agent Skill. Use when you want to capture a logging setup, CI/CD pipeline, observability stack, async messaging pattern, frontend component library, feature-flag…

Regression fixture for dry-run-replay header-skip drift.

A clean skill containing a git commit SHA, a UUID, and an embedded base64 image asset — none of which are secrets and none must trip the scanner.

Sets up structured logging for a service using a configurable log level and service name. Produces a logger.config file ready for ingestion by your logging pipeline. Works with any service that reads configuration from a flat config file.

A skill that accidentally contains an AWS access key — should be caught by the scanner.

A skill containing bare base64 secrets ending in equals-sign padding, which must not evade the entropy detector.

A clean skill with no secrets or sensitive identifiers. Uses only normal configuration values and parameterized placeholders.

A skill with compound credential keys like SECRET_KEY and CLIENT_SECRET that must be matched by the generic-credential detector.

Backward reasoning from desired outcome. Derives required conditions by asking 'What must be TRUE?'

Verify ALL artifacts against ALL constraints. Produces a verification matrix showing coverage and gaps

Light mode: 3-phase workflow for simple changes. Use instead of full workflow for bug fixes, small features, or quick iterations.

Generate parallel execution plan from constraint network. Identifies waves, critical path, and bidirectional dependencies

Initialize a constraint manifold for a feature. Creates .manifold/<feature>.json + .manifold/<feature>.md

Interview-driven constraint discovery across 5 categories (business, technical, UX, security, operational)

Surface and resolve constraint conflicts (trade-offs). Identifies direct conflicts, resource tensions, and competing requirements

Generate ALL artifact types from the constraint manifold in one coordinated run - code, tests, docs, runbooks, alerts

Walk a person through a subject, codebase, whitepaper, or complex system until they genuinely understand it. Use whenever the user wants to understand or learn something: "explain X", "walk me through this repo/paper", "help me understand how Y works",…

Design and build sound self-prompting loops — loops where a model iterates toward a goal across turns (draft→critique→revise, autonomous task loops, multi-agent orchestration, human-checkpointed loops). Use whenever the user wants to make a model "keep…

One-time SETUP that installs durable context management into a Claude Code project (or global ~/.claude): a bounded, scored, tiered cache that keeps working memory lean (anti-bloat) and rot-proof across compactions (anti-rot). Use when setting up, installing,…

One-time SETUP: installs a persistent SQLite-backed WORLD MODEL for a coding agent into a project (or global ~/.claude). Use when the user wants an agent to remember a codebase across sessions, detect contradictions, propose fixes, and improve correctness…

Run a code migration (language port, framework/SDK upgrade, service rewrite) as an operating model — rulebook + queues + reviewers + objective verification — instead of a pile of agent edits. Use when the user says "migrate this service", "port X to Y",…

Scaffold or extend a decision-oriented Astro + Starlight documentation handbook. Use when standing up a new handbook/knowledge-base site from scratch, or when adding a cluster (a group of topics) or a single topic page to an existing site that already follows…

Turn any Open Knowledge Format (OKF v0.1) bundle into a beautiful, browsable static website. Use when the user wants to publish, browse, view, share, or "make a site from" an OKF bundle / knowledge folder / agent-readable markdown wiki — e.g. "generate a site…

Adds a tmux-based agent cockpit with Zellij-like human ergonomics and Herdr-like agent supervision. Use when converting local tmux into a coding-agent workspace with menus, dashboard, pane navigation, per-agent status detection (Claude Code, Codex, Gemini,…

Autonomous codebase improvement loop inspired by Karpathy's autoresearch. Iteratively improves code quality, test coverage, performance, and architecture using multi-metric evaluation with diminishing returns detection. USE WHEN user wants to iteratively…

Autonomous codebase improvement loop inspired by Karpathy's autoresearch. USE WHEN user wants to iteratively improve a codebase, run autonomous code improvement, or apply the autoresearch pattern. Individual commands use /autoresearch directly.

This skill should be ALWAYS ACTIVE during software development. It automatically enforces the 10 Engineering Commandments when the user is writing code, planning features, creating specifications, designing architecture, or reviewing code. Auto-activates on…

This skill should be used when the user runs "/commandments", asks to "evaluate the codebase against engineering commandments", "assess engineering maturity", "run a commandments assessment", "check codebase maturity levels", "how mature is this codebase",…