agentleFS
Sign inStart free
quiczzquiczz/ship-safe-skills/skills/build-flow-map/SKILL.md

build-flow-map

Skill1 starsChanged 8 days agoMITView on GitHub

Map how an application is built and connected - languages, frameworks, entry points, build/run/test/deploy commands, environment variables, databases and external services, API routes, and CI/CD - and write it to docs/BUILD_FLOW.md with a diagram so any developer can onboard fast. Use when asked to document the build flow, track the code workflow, explain what frameworks the project uses, or onboard a new developer. Use proactively after structural changes (new framework, service, environment variable, route group) and when docs/BUILD_FLOW.md is missing or stale.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    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. Very specific:

    Runs on its own: after structural changes Claude updates docs/BUILD_FLOW.md, and offers to create it when it is missing.

    line 10
  2. Distinctive:

    Every command shown was found in a manifest, Makefile, or CI file (cite which).

    line 27
  3. Distinctive:

    Read-only on source code. The only file you create or change is the document (and optional CI snippet if asked).

    line 34
  4. Very specific:

    Do not print secrets. If .env (not .env.example) contains real values, use names only.

    line 35
  5. Distinctive:

    Do not claim a framework is used because it appears in a comment or a lockfile of a dependency.

    line 36

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Build Flow Map

Produces and maintains one document, docs/BUILD_FLOW.md, that answers a new developer's first questions: What is this made of? How do I run, test and ship it? What does it talk to? Where does a request go?

Runs on its own: after structural changes Claude updates docs/BUILD_FLOW.md, and offers to create it when it is missing.

Procedure

  1. Scan for facts (fast, no guessing):
    python3 <skill-dir>/scripts/scan_stack.py <project-root> --format markdown > /tmp/stack-scan.md
    The script reads manifests, lockfile names, Dockerfiles, CI workflows, .env.example, and source files to list: languages, frameworks, scripts/commands, environment variables (names only, never values), databases and services, HTTP routes, and a rough internal module graph.
  2. Verify and fill gaps by reading the code. The scan is heuristic. Open the entry points it found and confirm: how a request flows, where auth happens, what talks to the database, what runs in the background. Anything you cannot confirm goes under "Unverified" rather than being stated as fact.
  3. Write docs/BUILD_FLOW.md using references/build-flow-template.md. Keep it to what a developer needs; link to files instead of pasting code.
  4. Draw the connections. Include one Mermaid diagram of the runtime flow (browser/client, app, database, third parties, workers) and, for larger projects, one of the build/deploy pipeline. For prettier diagrams, hand the facts to a diagram skill if one is installed.
  5. Keep it current ("tracking"). Add a short "How to update this file" note at the bottom with the scan command. When you later add a framework, service, env var, route, or CI step, update the matching section in the same change. If the user wants it automatic, offer a CI step or pre-commit hook that runs the scan and fails when the file's "Last scanned" facts drift (see references/keeping-it-current.md).

What good output looks like

  • A newcomer can clone the repo and run it using only this file.
  • Every command shown was found in a manifest, Makefile, or CI file (cite which).
  • Environment variables are listed by name, with purpose and whether required. Never real values.
  • Connections are explicit: "API route POST /api/orders -> orders service -> Postgres table orders -> Stripe webhook".
  • Unknowns are labelled, not invented.

Rules

  • Read-only on source code. The only file you create or change is the document (and optional CI snippet if asked).
  • Do not print secrets. If .env (not .env.example) contains real values, use names only.
  • Do not claim a framework is used because it appears in a comment or a lockfile of a dependency; confirm it is a direct dependency or imported.

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.

Check your setup, free