Skip to content

Static Inventory

Static inventory mode scans test source files and emits structured metadata — class names, method names, line counts, and source-level annotations — without contacting an AI provider.

When to use this mode

  • You want a quick, zero-cost inventory of all test methods in a project without configuring an AI provider.
  • You are setting up a CI pipeline that needs a method inventory for delta comparison between builds.
  • You want to generate SARIF output for a GitHub Code Scanning dashboard without AI enrichment.
  • You are exploring a new project and want to understand its test structure before deciding on an AI strategy.
  • You plan to run AI enrichment later; static inventory gives you a baseline CSV to diff against.

Note: without AI enrichment, the output contains no ai_security_relevant, ai_tags, or ai_reason columns. Security classification requires API AI enrichment or the Manual AI workflow.

Basic scan

# CSV to stdout (default output)
./methodatlas src/test/java

# Save to file
./methodatlas src/test/java > inventory.csv

Sample output:

fqcn,method,loc,tags,display_name
com.example.AuthServiceTest,loginWithValidCredentials,12,,
com.example.AuthServiceTest,loginWithExpiredToken,8,security,
com.example.PaymentTest,chargeCard,15,,

Plain text output

./methodatlas -plain src/test/java

SARIF output for GitHub Code Scanning

./methodatlas -sarif src/test/java > results.sarif

Upload with the upload-sarif action:

- uses: github/codeql-action/upload-sarif@v3
  with:
    sarif_file: results.sarif

Without AI enrichment, all methods receive SARIF rule test-method and level none. Security-relevant classification requires Mode 2 or Mode 3.

Scan with metadata header

The -emit-metadata flag prepends scan provenance as comment lines before the CSV header — useful for archiving results with traceability information:

./methodatlas -emit-metadata src/test/java

# tool_version: 1.2.0
# scan_timestamp: 2025-04-09T10:15:30Z
# taxonomy: built-in/default
fqcn,method,loc,tags,...

Content hash fingerprinting

./methodatlas -content-hash src/test/java

Appends a SHA-256 hash of each class's AST representation to every record. When a class contains multiple test methods all of them share the same hash. Useful for incremental scanning and correlating results across pipeline runs.

Filtering which files are scanned

# Include integration tests alongside unit tests
./methodatlas -file-suffix Test.java -file-suffix IT.java src/test/java

# Recognise a custom test annotation (Java/Kotlin)
./methodatlas -test-marker Test -test-marker ScenarioTest src/test/java

The first occurrence of -file-suffix or -test-marker replaces the built-in defaults; subsequent occurrences append to the active set. The legacy -test-annotation flag is still accepted as an alias for -test-marker.

Scanning multiple source roots

./methodatlas module-a/src/test/java module-b/src/test/java

All roots are scanned in a single pass; the output is a single CSV. For monorepos where the same fully qualified class name appears in multiple modules, see Multi-root and monorepo scanning.

End-to-end scenario: baseline inventory before AI enrichment

A team is adopting MethodAtlas for the first time. They want a quick picture of how many test methods exist and which already carry a security tag — before committing to an AI provider subscription.

# Step 1: produce the baseline inventory
./methodatlas -emit-metadata -content-hash src/test/java > baseline.csv

# Step 2: count total test methods
echo "Total test methods: $(( $(wc -l < baseline.csv) - 1 ))"

# Step 3: count methods already tagged security
awk -F',' 'NR > 1 && $4 ~ /security/' baseline.csv | wc -l

Sample output from step 1:

# tool_version: 1.2.0
# scan_timestamp: 2026-04-29T09:00:00Z
# taxonomy: built-in/default
fqcn,method,loc,tags,display_name,content_hash
com.example.AuthServiceTest,loginWithValidCredentials,12,,, a3f1...
com.example.AuthServiceTest,loginWithExpiredToken,8,security,,b9c2...
com.example.PaymentTest,chargeCard,15,,,d4e5...

The team can see at a glance that loginWithExpiredToken is already tagged, while chargeCard is not. Later, when AI enrichment is configured, they diff the AI-enriched CSV against baseline.csv using the delta report to see what the AI reclassified.

For flags referenced here, see the CLI reference.