agentleFS
darrenCWJdarrenCWJ/agent-skills-scaffold/.claude/skills/orient/SKILL.md

orient

Skill0 starsChanged 3 months agoMITView on GitHub

Systematic onboarding to an unfamiliar project, repo, codebase, or resumed session — Phase 0 done properly. Use at the start of EVERY session in this scaffold, when entering a repo for the first time, when resuming after a gap, or before planning work in territory you haven't mapped.

27 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use retro, ui-ux-pro-max and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AWSSkill

    retro

    Conduct a retrospective on a coding session.

    Also in 3 projects that use this

  • StrayDragonSkill

    ui-ux-pro-max

    UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.

    Also in 3 projects that use this

  • AWSSkill

    ship

    Finalize dev work on an issue branch — update docs, push, and open a PR.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 25 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before planning, state briefly: the task restated in your own words.

    line 28

“Common” is measured: kept by at least 6 in 10 of the 25 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Orient

Work done on a wrong mental model is worse than no work — it must be found, understood, and undone. Orientation is cheap insurance. Never skip it because the task "seems simple"; simple tasks in misunderstood territory are how messes start.

Resume protocol (this scaffold's own state)

In order — later sources override earlier ones when they conflict:

  1. memory/MEMORY.md — read fully; open any entry whose description might touch the task.
  2. Newest file in docs/handovers/ — the successor protocol in the handover skill applies: verify claimed state, trust nothing.
  3. docs/decisions/ — scan titles; open ADRs adjacent to the task. Contradicting a standing decision without superseding it is a defect.
  4. docs/plans/ — is there an active (status: active/amended) plan this task belongs to?

New-territory protocol (unfamiliar repo/project)

  1. Read the self-descriptions first: README, AGENTS.md / CLAUDE.md, CONTRIBUTING, package manifests (package.json / pyproject.toml / go.mod...), CI config. These say what the project THINKS it is.
  2. Map the shape cheaply: directory tree 2 levels deep; where does source live, where do tests live, where does config live. Don't read files yet — map first, read second.
  3. Check the ground truth: git log --oneline -20 (what's actually being worked on), recently modified files, TODO/FIXME density. The gap between self-description and ground truth is where surprises live.
  4. Find the conventions: naming style, error-handling pattern, test structure — from 2–3 representative files. New work must look native, not transplanted.
  5. Locate the verification path: how does THIS project prove things work (test command, build, lint)? Run it BEFORE changing anything — a pre-existing red baseline discovered after your changes looks like your fault and costs an hour of misdirected debugging.

Output (the orientation statement)

Before planning, state briefly: the task restated in your own words; what exists that's relevant; the conventions that apply; the verification path; unknowns split into "will investigate" vs "must ask user". If any unknown blocks planning, batch ALL user questions into one ask.

Budget

Orientation is minutes, not hours — timebox it (5–15 min of reading for most tasks). Going deeper than needed is procrastination wearing a safety vest; the plan phase will drive targeted deep-dives where they matter.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.