By rengwu
Plan and execute large, ambiguous efforts by mapping them as investigation tickets under .plan/, resolved one per session. Companion skills guide structured interviews, domain modeling, throwaway prototyping, research from primary sources, code review against spec, and session handoffs.
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 the domain model.
A relentless interview to sharpen a plan or design — one question at a time until every branch of the decision tree is resolved.
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.
Investigate a question against high-trust primary sources and capture the findings as a cited Markdown file. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated.
Plan a huge chunk of work — more than one agent session can hold — as a shared map of investigation tickets, and resolve them one at a time until the way to the destination is clear.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A read-only CLI and viewer for
wayfinder
maps — the markdown planning memory an agent leaves under .plan/<effort>/ as it charts
a large effort as a graph of investigation tickets. The wayfinder method and its skills
were created by Matt Pocock; this repo adapts them and
adds the tooling.
wayfinder-maps status <dir> — what is resolved, what is in flight, and what is ready
to claim (the frontier: open, unclaimed, every blocker resolved).wayfinder-maps lint <dir> — does the map still tell the truth?wayfinder-maps serve / app — the map as a star-map, in a browser or a native window
(design notes).$ wayfinder-maps status ../expensif/.plan/daily-timeline
Daily Timeline — continuous days, empty days included
6 resolved · 0 claimed · 4 open · 0 out of scope
Frontier — ready to claim, first by number wins:
05 Should HandleDaily's two branches converge grilling
06 Contain the day-card chrome drift grilling
07 The infinite-scroll island's contract grilling
10 Test strategy for the date-indexed timeline grilling
Undermined — resolved on a premise that later changed:
02 Window size and how older days load broken by 08
Fog: 5 patches, 1 anchored to a ticket
/plugin marketplace add rengwu/wayfinder-maps
/plugin install wayfinder-maps@wayfinder-maps
/wayfinder-maps then works in every project; the map lands in that project's .plan/. The
plugin bundles wayfinder-maps plus the four skills it invokes by name — grill-me,
research, prototype, domain-modeling. All five were originally authored by Matt
Pocock in mattpocock/skills; the copies under
skills/ are MIT-licensed adaptations.
Copy skills/ into your project and tell your agent to read
skills/wayfinder-maps/SKILL.md and follow it; the skills it names live alongside.
You may include any of the extras under skills/.optional/ too —
skills/README.md describes the full set.
libwebkit2gtk-4.1, preinstalled on most desktops, one
sudo apt install libwebkit2gtk-4.1-0 otherwise. status and lint need nothing
installed at all, even headless. The macOS and Linux archives also carry two
double-clickable launchers — wayfinder-app opens the native window, wayfinder-web
starts the server and opens your browser — no terminal needed (.command on macOS).go install github.com/rengwu/wayfinder-maps/cmd/wayfinder-maps@latestSkill and binary are independent — each just reads the on-disk contract.
The map is shared memory: sessions orient to it before choosing work, and when it drifts
it lies to them silently. The skill makes most drift unrepresentable and verifies each
session's delta, no tool assumed. What a delta check cannot see is an edit made outside
the protocol, or two parallel sessions taking the same ticket number. lint is fsck
for those: it runs after the fact and re-establishes the base the next session rests on.
The contract is the skill's local-markdown adapter,
TRACKER-MARKDOWN.md — the skill itself is
tracker-agnostic method. This tool implements that adapter; any other reader should
target the same file. In summary: structure the facts, leave the prose alone. A ticket's
type and edges have exactly one correct value a machine can check. Its Question and
Answer, the map's Destination and Notes, and each one-line gist in
Decisions-so-far are lossy prose, never parsed.
A ticket is .plan/<effort>/tickets/NN-<slug>.md:
---
type: research | prototype | grilling | task
blocked_by: [02, 03] # [] when none
claimed_by: <session id> # while a session holds it
claimed_at: 2026-07-10T09:00:00Z # RFC 3339, alongside claimed_by
undermined_by: [08] # optional — resolved on a premise 08 later broke
assets: [../assets/x.html.approved] # optional
---
# <Ticket title>
## Question
…
## Answer # its presence _is_ the resolution
…
There is no status: field — every value it could hold is already written in the file,
and a second copy is free to go stale:
npx claudepluginhub rengwu/wayfinder-mapsHarness-native ECC plugin for engineering teams - 67 agents, 278 skills, 94 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Binary reverse engineering, malware analysis, firmware security, and software protection research for authorized security research, CTF competitions, and defensive security
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications
Next.js development expertise with skills for App Router, Server Components, Route Handlers, Server Actions, and authentication patterns
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use