faf-python-sdk

Persistent Project Context for Python. Parse, validate, score.

FAF defines. MD instructs. AI codes.

The foundation other Python FAF tools build on. If you're building MCP servers, CI validators, or any Python tool that needs to understand project context, start here.

Media Type: application/vnd.faf+yaml (IANA registered)

What's New in v1.2.0 — The Dart Edition

Adds detect_dart_project(): content-aware Dart/Flutter detection from a pubspec.yaml (Flutter app vs package · Dart MCP / backend / CLI / library), reproducing faf-cli's engine byte-for-byte — 20 shared fixtures, parity-tested.

The SDK's first detection capability — it now reads a pubspec.yaml and tells you what kind of Dart project it is, not just parse/validate/score. The detection KNOWLEDGE is vendored byte-identical from faf-cli (the single source) and ships in the wheel:

from faf_sdk import detect_dart_project

d = detect_dart_project(".")
print(d.app_type, d.framework)   # e.g. "mobile" "Flutter"

What's New in v1.1.0

Mk4 Championship Scoring Engine — the same 33-slot scoring algorithm used by the Rust compiler and TypeScript CLI, now in Python. Same slots, same formula, same scores. Every FAF tool in every language now agrees on what 100% means.

• score_faf() — Mk4 scoring with 21-slot Base or 33-slot Enterprise tiers
• 100% parity with faf-wasm-sdk (Rust) and faf-cli (TypeScript)
• 3 crash bugs fixed (malformed YAML, null project fields)
• 175 tests including 88 WJTTC championship-grade tests (concurrency, adversarial input, security)

Why this matters: If you're building on FAF in Python — MCP servers, Gemini extensions, CI pipelines — your scores now match every other FAF tool exactly. No more "it scored 85% in the CLI but 60% in Python." One engine, one truth.

v1.1.2 is a patch release — package description aligned with the canonical "Persistent project context for Python" framing. CHANGELOG.md added. No code changes.

Installation

pip install faf-python-sdk

Quick Start

from faf_sdk import parse_file, score_faf

# Parse a .faf file
faf = parse_file("project.faf")
print(f"Project: {faf.project_name}")

# Score it with the Mk4 engine
with open("project.faf") as f:
    result = score_faf(f.read())

print(f"Score: {result.score}% {result.tier}")
print(f"Slots: {result.populated}/{result.total} populated")

FAF defines. MD instructs. AI codes.

Mk4 Scoring

The Mk4 engine scores .faf files by checking 21 universal slots (project metadata, human context, tech stack). Each slot is Populated, Empty, or Slotignored. The score is the percentage of active slots that are populated.

from faf_sdk import score_faf, LicenseTier

# Base scoring (21 slots)
result = score_faf(yaml_content)
print(result.score)      # 0-100
print(result.tier)       # Trophy/Gold/Silver/Bronze/Green/Yellow/Red
print(result.populated)  # slots with real data
print(result.active)     # total minus slotignored
print(result.slots)      # per-slot breakdown

# Enterprise scoring (33 slots — adds monorepo/infra)
result = score_faf(yaml_content, LicenseTier.ENTERPRISE)

Placeholder rejection: Values like "null", "unknown", "n/a", "Describe your project goal" are detected and scored as Empty — not Populated.

Slotignored: Set any slot to slotignored to exclude it from scoring. A backend-only project can mark frontend: slotignored and still reach 100%.

Parsing

from faf_sdk import parse, parse_file, stringify

# Parse from string or file
faf = parse(yaml_content)
faf = parse_file("project.faf")

# Typed access
print(faf.data.project.name)
print(faf.data.project.goal)
print(faf.data.stack.backend)
print(faf.data.human_context.who)

# Raw dict access
print(faf.raw["project"]["goal"])

# Convert back to YAML
yaml_str = stringify(faf)

Validation

from faf_sdk import validate

result = validate(faf)

if result.valid:
    print(f"Valid! Score: {result.score}%")
else:
    print("Errors:", result.errors)

print("Warnings:", result.warnings)

File Discovery

from faf_sdk import find_faf_file, find_project_root

# Find project.faf (walks up directory tree)
path = find_faf_file("/path/to/src")

# Find project root by markers (package.json, pyproject.toml, .git, etc.)
root = find_project_root()

API Reference

Function	Returns	Description
score_faf(yaml, tier?)	Mk4Result	Mk4 score (21 or 33 slots)
parse(content)	FafFile	Parse YAML string
parse_file(path)	FafFile	Parse from file path
validate(faf)	ValidationResult	Structure validation + warnings
stringify(data)	str	Convert back to YAML
find_faf_file(dir?)	str | None	Find project.faf in tree
find_project_root(dir?)	str | None	Find project root

FAF Ecosystem

Package	Platform	Registry
faf-python-sdk	Python foundation	PyPI
gemini-faf-mcp	Google Gemini	PyPI
claude-faf-mcp	Anthropic	npm + MCP #2759
grok-faf-mcp	xAI	npm
faf-cli	CLI	npm

If faf-python-sdk has been useful, consider starring the repo — it helps others find it.

Links

• Site: faf.one
• IANA Registration: application/vnd.faf+yaml
• Gemini MCP: gemini-faf-mcp

License

MIT