agentleFS
DeevsDeevsDeevsDeevs/agent-system/skills/codebase-orientation/SKILL.md

codebase-orientation

Skill41 starsChanged 6 days agoApache-2.0View on GitHub

Map an unfamiliar code area before editing: 'orient me', 'zoom out', 'where does this fit', multi-file or subsystem relationships.

4 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions4 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 4

    1. proflead/codex-skills-library★ 153same name
    2. adamentwistle/fable-skills★ 98same name
    3. HyperVon/agent-guidance-kit★ 1same name
    4. Bridev04/clauderail★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository7 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 2

    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:

    Stop and summarize: map plus recommended next step.

    line 28
  2. Distinctive:

    No implementing or root-causing during orientation, no giant repo maps, no asking the user for cheaply readable facts, no trusting docs …

    line 56

“Common” is measured: kept by at least 6 in 10 of the 4 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

Codebase Orientation

Build the smallest accurate map that makes the next action safe. Orientation, not diagnosis or architecture critique — with a concrete failing symptom, orient only far enough to find the area, then switch to diagnose.

Contract

Bound the map before exploring; ask one scope question if scope is unclear. Never map the whole repo by default.

Target: [feature/module/crate/error path]
Scope: [repo root + allowed dirs]
Question: [what the map must answer]
Stop when: [next safe action is clear]

Operating loop

  1. Find repo shape and rules: project instructions, workspace layout, entrypoints, docs, tests, scripts.
  2. Trace the relevant path from user-visible trigger or public API to the code that matters.
  3. Name core concepts: types, traits, modules, services, state, invariants.
  4. Find callers/callees — enough to understand impact, not every reference.
  5. Find validation hooks: narrow build/test/check commands and fixtures.
  6. Stop and summarize: map plus recommended next step.

Bounded searches only: rg, find, package metadata, obvious entrypoints. Source-of-truth order: source/tests/config/CI beat project docs; project docs beat chains. Mark uncertainty instead of smoothing conflicts.

Tools: read files directly, shell for bounded searches, an explore subagent only when the area spans enough files that a bounded read-only map helps (explicit cwd, paths, output shape; continue other mapping work while it runs), a small Mermaid diagram only when it beats prose, a chain save only when the map is useful future context.

Rust checklist (inspect only what is relevant)

Workspace/crate boundaries (Cargo.toml, cargo metadata --no-deps, features) · command conventions (README, AGENTS.md, Justfile, CI) · entrypoints (lib.rs, main.rs, mod.rs) · public API and module tree · key structs/enums/traits · state sharing (Arc, locks, caches, globals) · async boundaries, channels, retries, timeouts · error types and Result flow · feature flags and cfg gates · tests/fixtures and narrow commands (cargo check -p, cargo test -p <crate> <filter>). No full-workspace checks unless narrower ones are impossible.

Output

Compress small orientations; skip empty headings. Include file paths; distinguish facts from inferences.

What this area does
Main files/modules — path, why it matters
Entrypoints
Important types/contracts
Call/data flow
State and invariants
Validation hooks — command, what it proves
Risks before editing
Recommended next step

Pitfalls

No implementing or root-causing during orientation, no giant repo maps, no asking the user for cheaply readable facts, no trusting docs over source, no hiding uncertainty. A list of files is not an orientation — explain why each file matters.

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.