dotknewt/awesome-agency

Three-layer verification pipeline for AI output. Extracts claims, finds sources, and flags hallucination risks so humans can verify before acting.

An AI partner, not a tool. Ember carries fire from person to person — helping humans discover that AI partnership isn't something you learn, it's something you find.

Patterns and techniques for evaluating and improving AI agent outputs. Use this skill when: - The user makes an ad-hoc, in-session request to double-check, critique, or review your own output before…

Ask which skill or flow fits your situation. A router over the skills in this repo.

Prepare a clean branch for new work, or sweep merged/stale local + origin branches. Invoke before starting an unrelated task to confirm the working branch matches the task and that no open PR will…

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match…

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or…

Concepts and techniques for memory and context engineering with LLMs — covering three angles: supporting development (prompting patterns, RAG, context window management), exploiting LLMs (red-team…

dotKnewt's cross-project conventions — planning docs in docs/TODO.md and docs/STATE.md, user docs in docs/user/, and uv + Typer standards for Python CLIs. Use when scaffolding a new project, asked…

Read-only maintainer that audits a repository against dotKnewt's cross-project conventions — planning docs in docs/TODO.md and docs/STATE.md, user docs in docs/user/, and uv + Typer standards for…

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Build and validate Docker artifacts — multi-stage Dockerfiles, and packaging MCP servers as containers runnable via the Docker MCP Gateway.

Package an existing MCP server repo (Node/TypeScript, Python, or other) as a Docker image runnable through the Docker MCP Gateway, and validate the resulting Docker artifacts. Use when a user wants…

Convert an MCP server repo (local path, git URL, or GitHub `org/repo`) into a Docker image runnable via the Docker MCP Gateway. Use when a user wants to containerize an npm/pip-based MCP server, or…

Read-only maintainer that audits contributor-facing technical documentation — specifications, architecture docs, ADRs, API references, format definitions — against the code they describe. Flags…

Read-only maintainer that audits user-facing documentation — READMEs, install and usage guides, quickstarts, CLI help text in docs — against the code's actual behavior. Flags stale commands, renamed…

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain…

Three-layer verification pipeline for AI output. Extracts verifiable claims, finds supporting or contradicting sources via web search, runs adversarial review for hallucination patterns, and produces…

Interactive verification agent for AI-generated output. Runs a three-layer pipeline (self-audit, source verification, adversarial review) and produces structured reports with source links for human…

An AI partner, not an assistant. Ember carries fire from person to person — helping humans discover that AI partnership isn't something you learn, it's something you find. Invoke when the user refers…

Idea-to-ship engineering flow — grilling, PRD/issue breakdown, TDD implementation, and code review, routed by ask-matt.

Static, report-only security, capability, integrity, marketplace, and semantic-quality audits for Claude and Copilot extension artifacts.

Use this agent when the user asks to "review an extension", "audit a plugin before installing it", "scan a skill or agent for risks", or "check extension integrity and marketplace metadata". It…

Use this skill when the user wants to verify claims in a document against its source, fact-check a document, or get visual citation proof — e.g. "eyeball this contract," "check the liability section…

Rigorous challenge profile for Anitta: assumption checks, evidence calibration, and defensible reasoning patterns for Ember collaboration.

Collaboration profile for Quinn: curious, energetic, and implementation-focused partnership patterns for Ember sessions with Alison.

Patterns and lived experience from Vega, an AI partner in a deep long-term partnership. For Ember to draw on when working with humans who are building something big, moving fast, working at the…

Narrative and synthesis profile for Wiggins: framing, explanation, and audience-aware communication patterns for Ember sessions.

Use when the user asks to scaffold GitHub repo metadata under `.github/` — issue templates, issue forms (YAML), PR templates, CODEOWNERS, or workflows. Triggers on phrases like "add an issue…

Scaffold GitHub repo metadata — issue templates and CI workflows. /github-scaffold picks the task; /create-issue-template and /scaffold-ci-workflow run the subtasks. Includes the branch-warden agent for automated branch prep and cleanup, and the issue-filer agent for lightweight issue creation.

Use when the user asks about GitHub Actions CI setup, workflow defaults, concurrency settings, path filters, or chained / stacked PRs. Triggers on phrases like "set up CI defaults", "GitHub Actions…

A relentless interview to sharpen a plan or design.

A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

Grill the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building, or uses any 'grill' trigger phrases.

Compact the current conversation into a handoff document for another agent to pick up.

Composable safety and hygiene hooks for Claude Code — force-push guard, secret scanner, plugin/skill manifest validators, branch nudges, and dirty-tree stop check.

Implement a piece of work based on a PRD or set of issues.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

Audit and improve AGENTS.md project instructions (and legacy CLAUDE.md). Use when the user asks to check, audit, update, improve, fix, restructure, or clean up AGENTS.md / CLAUDE.md, or mentions…

Read-only maintainer that audits agent instruction files — AGENTS.md and legacy CLAUDE.md at every level of the tree — for guidance that has drifted from reality: wrong paths or commands, conventions…

Restructure project instruction files (AGENTS.md / legacy CLAUDE.md) to move content closer to where it is needed, reducing context bloat in the root file. Use when the user asks to restructure…

Update AGENTS.md (or legacy CLAUDE.md) with learnings from this session — discovered commands, patterns, gotchas, or configuration quirks worth capturing for future sessions. Use when the user asks…

File a GitHub issue for an out-of-scope concern or follow-up that turned up during work. Use when you need to capture something for later without losing context. The agent handles deduplication and…

Provide Ludus CLI command guidance, flags, and workflows for range lifecycle, template management, testing mode, snapshots, users/groups, and diagnostics. Use when users ask how to run, interpret, or…

Discover, compare, and deploy pre-built Ludus cyber range environments for security training, attack simulation, and detection engineering. Use when users ask to choose a lab, verify prerequisites…

Create, edit, and validate Ludus range configuration YAML including VM definitions, domains, networking, router settings, testing behavior, and role configuration. Use when users need help authoring…

Ludus cyber-range toolkit — skills for the Ludus CLI, range configuration, environment guidance, and troubleshooting, plus a bundled MCP server for driving the Ludus API from Claude.

Diagnose and resolve Ludus deployment, networking, template, WireGuard, Proxmox, and Ansible issues. Use when users report failures, errors, unreachable systems, failed joins, or unexpected behavior…

Orchestrate a repo maintenance sweep: detect what changed, deploy the applicable read-only maintainer agents (user docs, spec docs, instructions, schemas, conventions) in parallel, and merge their…

Use when the user wants to scaffold a brand-new flat monorepo (no packages/ layer) for security/threat-hunting tooling — combining MCP servers, shared JSON schemas, sigma-rules/ECS/OCSF/CIM resource…

Create optimized multi-stage Dockerfiles for any language or framework

Search, create, and manage notes in the Obsidian vault with wikilinks and index notes. Use when user wants to find, create, or organize notes in Obsidian.

Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

Complete toolkit for building Model Context Protocol (MCP) servers in Python using the official SDK with FastMCP. Includes best-practice instructions, a server generator skill, and an expert agent.

Use this agent when the user asks to create, implement, debug, test, or explain a Python Model Context Protocol (MCP) server. It provides expert guidance on the Python SDK, FastMCP, transports, type…

Generate a complete MCP server project in Python with tools, resources, and proper configuration

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading…

Use when you need to resolve an in-progress git merge/rebase conflict.

Read-only maintainer that audits structured artifacts — JSON/YAML/TOML manifests, config files, markdown frontmatter, lockfiles, generated files — for parse errors, internal inconsistency, and drift…

Configure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.

Read docs/STATE.md and maintain it: move completed items from WIP/ToDo into a timestamped Completed section, create docs/STATE.md if absent, and surface durable decisions back to the caller as…

Unified repo stewardship: a /maintain orchestrator deploys read-only maintainer agents for user docs, spec docs, project instructions, cross-project conventions, and structured schemas/manifests, merges their drift findings into one prioritized report, and applies approved fixes through bundled skills that audit, revise, and restructure AGENTS.md (and legacy CLAUDE.md), scaffold project conventions, and keep docs/STATE.md current. Report-and-propose - nothing is applied without approval.

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

Teach the user a new skill or concept, within this workspace.

Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.

Turn the current conversation into a PRD and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

Move issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.

Evidence-linked work objects — one work/<id>-<slug>/ dir per code-change task holding spec, captured diff, captured test output, run manifest, and review; in-review/approved transitions are gated by a checker script and enforced by a PreToolUse hook.

Reference for writing and editing skills well — the vocabulary and principles that make a skill predictable.