agentleFS
chuckplayerchuckplayer/claude-agent-pack/skills/onboard/SKILL.md

onboard

Skill8 starsChanged 35 days agoMITView on GitHub

Generate a structured orientation for a codebase. Reads the project, memory, and conventions to produce a mental model: architecture, key entry points, data flow, and known gotchas. Use when joining a project, returning after time away, or onboarding a new contributor. Trigger this when someone says: orient me, explain this codebase, how does this project work, I am new here, give me an overview, help me understand the architecture, catch me up. Do NOT use to document conventions — use /conventions instead. Do NOT use to plan a task — use /plan instead.

  • Reads credentials

169 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use soak-test, team-live-ops and 10 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.
  • lilangMaxSkill

    soak-test

    Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.

    Also in 9 projects that use this

  • lilangMaxSkill

    team-live-ops

    Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.

    Also in 9 projects that use this

  • HK-hubSkill

    teach-impeccable

    One-time setup that gathers design context for your project and saves it to your AI config file. Run once to establish persistent design guidelines.

    Also in 19 projects that use this

  • CoralGameSkill

    project-stage-detect

    Automatically analyze project state, detect stage, identify gaps, and recommend next steps based on existing artifacts.

    Also in 15 projects that use this

  • CoralGameSkill

    team-level

    Orchestrate level design team: level-designer + narrative-director + world-builder + art-director + systems-designer + qa-tester for complete area/level creation.

    Also in 11 projects that use this

  • +5 more from lilangMax
  • +2 more from CoralGame

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 157 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the following in parallel:

    line 12
  2. Very specific:

    All files in memory//*.md (skip any with status: superseded or status: archived)

    line 17
  3. Very specific:

    If a current repo-map.md covered the structure in step 1, use it instead of re-deriving and move to step 3.

    line 23
  4. Distinctive:

    After the orientation, note:

    line 96
  5. Distinctive:

    No README or CONVENTIONS.md: Do not invent documentation.

    line 103
  6. Distinctive:

    This skill is read-only: Do not create, modify, or delete any files during onboarding.

    line 106

“Common” is measured: kept by at least 6 in 10 of the 157 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

Onboard

Produce a structured orientation for this codebase. This skill is read-only — it makes no changes.

1. Collect project context

Read the following in parallel:

  • README.md (or readme.md) at the project root
  • docs/CONVENTIONS.md if it exists
  • CLAUDE.md at the project root if it exists
  • All files in memory/**/*.md (skip any with status: superseded or status: archived)

If memory/architecture/repo-map.md exists, read it first — it is a cached directory-level map of the codebase. Check its verified-at-commit stamp: run git diff --name-only <that-sha>..HEAD to see whether the tree has drifted. If it is current, use it as the backbone of the orientation and skip the exhaustive re-exploration in step 2 (spot-check only). If it is stale, note that and recommend /repo-map refresh, then fall back to full exploration for the drifted areas.

2. Explore the codebase structure

If a current repo-map.md covered the structure in step 1, use it instead of re-deriving and move to step 3. Otherwise use Glob and Grep to map the project. Tailor the exploration to what you find — not all projects have all layers:

Structure:

  • Top-level directories and their apparent purpose
  • Solution/project files (.sln, package.json, pyproject.toml, etc.)
  • Key configuration files (appsettings.json, .env.example, vite.config.ts, etc.)

Entry points:

  • API controllers or route definitions
  • Background jobs or scheduled tasks
  • Frontend pages or top-level components
  • Main program entry point

Data layer:

  • Database context or ORM configuration
  • Migration files (count and rough date range)
  • Key entities or models

Frontend (if present):

  • Component structure
  • State management (stores, composables)
  • API client layer

3. Identify key patterns

From what you've read, identify:

  • The architectural style (layered, CQRS, modular monolith, microservices, etc.)
  • Naming conventions actually in use (may differ from CONVENTIONS.md)
  • Testing approach (unit, integration, e2e — what framework, where tests live)
  • Anything flagged in memory as a known issue or non-obvious constraint

4. Produce the orientation document

Depth target: Aim for the level of detail a developer new to this project could use to make their first commit within two hours. If a section would require deep explanation to be useful, flag it in step 5 instead of expanding it here — a pointer to the right file is more useful than an incomplete explanation.

Output a structured orientation using this format:

# Project Orientation: <project name>

## What this project does
<1–3 sentences>

## Architecture overview
<Describe the major layers and how they connect. Include a simple text diagram if helpful.>

## Key entry points
<List the most important files a new contributor would need to find first.>

## Data flow
<Trace a representative request or operation from entry point to data store and back.>

## Where to find things
| Concern | Location |
|---------|----------|
| API endpoints | ... |
| Business logic | ... |
| Database models | ... |
| Tests | ... |
| Frontend components | ... |

## Conventions and standards
<Summarize the most important rules from CONVENTIONS.md and memory. Flag any gaps or inconsistencies between documented conventions and actual code.>

## Known gotchas
<List anything from memory/known-issues/ or from code patterns that would surprise a newcomer.>

## Suggested first tasks
<2–3 low-risk starting points for someone new to this codebase.>

5. Flag gaps

After the orientation, note:

  • What you could not determine from reading (e.g., no README, no CONVENTIONS.md, no memory files)
  • Any apparent inconsistencies between documentation and actual code
  • Areas where a human familiar with the project should fill in missing context

Gotchas

  • No README or CONVENTIONS.md: Do not invent documentation. If these files are missing, say so clearly in step 5 and suggest running /setup-project and /conventions to create them. The orientation should reflect what actually exists.
  • Memory files out of sync with code: Memory records what was true at the time they were written. Always verify a memory claim against the current code before stating it as fact. If a memory file says "we use X" but the code uses Y, report the discrepancy rather than picking one.
  • Depth target mismatch: The goal is "first commit in two hours," not a comprehensive architecture document. If you find yourself writing three paragraphs about a single module, that's a pointer to a reference file, not inline content. Keep it navigable.
  • This skill is read-only: Do not create, modify, or delete any files during onboarding. If you identify something that should be documented, recommend /conventions or /memory-audit — do not write it yourself.

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.