Configuration

Reference

Use impact-gate.config.json to pin analysis defaults, route the tool to the right tests root, and make CI behavior predictable across pull requests, release branches, and local runs.

Config purpose

Use config when you want repeatable team defaults

Auto-detection is great for onboarding, but config is what makes CI, release branches, and local runs behave the same way across the team.

File names

The CLI looks for these files automatically

impact-gate.config.json .impact-gate.config.json

Top-Level git impact pipeline policy Profiles

Create impact-gate.config.json (or .impact-gate.config.json) in your project root. The CLI auto-discovers it by walking upward from the current directory.

Full Config Example

{
  "path": ".",
  "profile": "default",
  "testsRoot": "./e2e-tests",
  "mode": "impact",
  "framework": "auto",
  "git": {
    "since": "origin/main"
  },
  "impact": {
    "dependencyGraph": {
      "enabled": true,
      "maxDepth": 3
    },
    "traceability": {
      "enabled": true
    },
    "aiFlow": {
      "enabled": true,
      "provider": "anthropic"
    }
  },
  "pipeline": {
    "enabled": false,
    "scenarios": 3,
    "outputDir": "specs/functional/ai-assisted",
    "mcp": false
  },
  "policy": {
    "enforcementMode": "advisory",
    "blockOnActions": ["must-add-tests"]
  }
}

Field Reference

Top-Level

Field	Type	Default	Description
path	string	.	Project root directory
profile	string	default	Analysis profile: default or strict
testsRoot	string	auto-detected	Path to tests directory
mode	string	impact	Default analysis mode
framework	string	auto	Test framework (playwright, cypress, pytest, supertest, selenium, auto)

git

Field	Type	Default	Description
since	string	auto-detected	Git ref for diff base (e.g., origin/main)

impact

Field	Type	Default	Description
dependencyGraph.enabled	boolean	true	Enable static reverse-dependency analysis
dependencyGraph.maxDepth	number	3	Max depth for transitive impact traversal
traceability.enabled	boolean	true	Use CI execution data for file-to-test mapping
aiFlow.enabled	boolean	true	Enable LLM-powered flow mapping
aiFlow.provider	string	auto	LLM provider for AI enrichment

pipeline

Field	Type	Default	Description
enabled	boolean	false	Enable test generation pipeline
scenarios	number	3	Number of test scenarios to generate per gap
outputDir	string	—	Directory for generated specs
mcp	boolean	false	Use Playwright MCP server for generation

policy

Field	Type	Default	Description
enforcementMode	string	advisory	advisory, warn, or block
blockOnActions	string[]	[]	Actions that trigger blocking (run-now, must-add-tests, safe-to-merge)

Profiles

Profile

default

Standard analysis behavior for most repositories.

Profile

strict

Uses stricter handling for heuristic-only mappings and more opinionated analysis defaults when you want tighter gating.

Framework auto-detection is separate from profiles. The CLI can auto-detect Playwright, Cypress, pytest, supertest, and Selenium usage from project files and dependencies.

Precedence

CLI flags override config file values, which override auto-detected defaults.