brokkai/brokk-skills/brokk-untested-paths

Name: brokk-untested-paths
Author: brokkai

Claude Code and agent skill by brokkai/brokk-skills

Trust tier 3

community

Source check pending

Checking that the SKILL.md file is still in the repository.

Use when asked to find untested code, missing test coverage, or pathways with no tests in a codebase - maps source files and functions against test files to surface gaps

Install

npx vskill@latest install brokkai/brokk-skills/brokk-untested-paths

SKILL.md

brokk-untested-paths

Overview

Find code paths with no test coverage by cross-referencing source files against test files, detecting test framework conventions, checking for coverage reports, and identifying critical functions that appear nowhere in the test suite.

Prerequisites

rg --version 2>/dev/null || echo "rg MISSING"

Step 1 — Detect Test Framework & Convention

# Find test files by common naming conventions
rg --files | rg -i "(test|spec)\.(py|js|ts|java|go|rb|cs|php|rs)"
rg --files | rg -i "(test|spec)s?/"
rg --files | rg "_test\.(go|py|rs)$"

Language	Test file pattern	Framework markers
Python	`test_.py`, `_test.py`	`import pytest`, `import unittest`, `def test_`
JS/TS	`.test.ts`, `.spec.ts`	`describe(`, `it(`, `test(`, `expect(`
Java	`Test.java`, `Tests.java`	`@Test`, `import org.junit`
Go	`*_test.go`	`func Test`, `func Benchmark`
Ruby	`_spec.rb`, `test_.rb`	`describe`, `it`, `expect`
C#	`Tests.cs`, `Test.cs`	`[Test]`, `[Fact]`, `[TestMethod]`
Rust	same file, `#[cfg(test)]`	`#[test]`
PHP	`*Test.php`	`extends TestCase`

Note the conventions found. This shapes Step 3.

Step 2 — Check for Existing Coverage Reports

Before doing manual analysis, look for pre-generated coverage data:

rg --files | rg -i "(coverage|lcov|\.coverage|htmlcov|coverage\.xml|clover\.xml|jacoco)"
# Common locations
# Python: .coverage, htmlcov/, coverage.xml
# JS: coverage/lcov.info, coverage/lcov-report/
# Java: target/site/jacoco/index.html
# Go: coverage.out
# Ruby: coverage/.resultset.json
# Rust: tarpaulin-report.html

If a coverage report exists, read it and extract:

Overall line/branch coverage %
Files with 0% coverage
Files with <50% coverage

Report these directly — they are the answer. Still do Step 3 to find structurally untested paths the report may miss.

Step 3 — Map Source Files to Test Files

Build two lists:

Source files (exclude test dirs, vendor, node_modules, generated):

rg --files --glob "!*test*" --glob "!*spec*" --glob "!vendor/*" \
   --glob "!node_modules/*" --glob "!dist/*" --glob "!build/*" \
   | rg "\.(py|js|ts|java|go|rb|cs|php|rs)$"

Test files:

rg --files | rg -i "(test|spec)\.(py|js|ts|java|go|rb|cs|php|rs)|_test\.(go|py|rs)"

For each source file src/foo/bar.py, check whether a matching test file exists:

tests/foo/test_bar.py ✓
tests/foo/bar_test.py ✓
No match ✗ → untested file

List all source files with no corresponding test file. These are candidates — confirm below.

Step 4 — Find Untested Functions

For each file flagged in Step 3, list its public functions/methods:

# Python — public functions (not starting with _)
rg -n "^def [^_]|^    def [^_]" src/foo/bar.py
 
# JS/TS — exported functions
rg -n "^export (async )?function|^export (const|let) \w+ = " src/foo/bar.ts
 
# Java — public methods
rg -n "public \w+ \w+\(" src/Foo.java
 
# Go — exported functions (capital first letter)
rg -n "^func [A-Z]" foo.go
 
# Ruby — public methods (not private/protected)
rg -n "^\s*def [^_]" foo.rb

Then check whether each function name appears anywhere in the test suite:

rg -l "function_name" tests/

If the function name does not appear in any test file → untested function.

Step 5 — Identify High-Risk Untested Paths

Not all untested code is equal. Flag as critical if the untested code:

Is in files touched by recent commits: git log --oneline -30 -- <file>
Handles auth, payments, data mutation, file I/O, external APIs
Contains error handling (except, catch, rescue) — edge cases are rarely tested
Is in a file with no test at all (not just low coverage)

# Find recently changed untested files
git log --since="90 days ago" --name-only --format="" | sort -u \
  | rg "\.(py|js|ts|java|go)$" > recently_changed.txt
# Cross-reference with untested file list

Output Format

markdown

## Untested Paths Report
 
### Coverage Summary (if report found)
-Overall: 34% line coverage
-Files with 0% coverage: 12
-Files with <50% coverage: 28
 
### Source Files With No Test File
| File | Public Functions | Recently Changed |
|------|-----------------|-----------------|
| src/auth/token.py | 6 | yes (14 days ago) |
| src/billing/invoice.py | 4 | no |
 
### Untested Functions (in partially-tested files)
| Function | File | Risk |
|----------|------|------|
| validate_payment() | src/billing/payment.py:88 | HIGH — handles money |
| refresh_token() | src/auth/token.py:42 | HIGH — auth path |
 
### Test Framework Detected
-pytest (Python)
-jest (JS)
 
### Coverage Tool Detected
-.coverage found — run `coverage html` to view report

Windows Notes

rg --files and rg -l work natively
Replace shell pipeline sort -u | rg with: collect rg output, deduplicate manually
git log works natively on Windows with Git for Windows

Common Mistakes

Mistake	Fix
Reporting private/internal functions as untested	Focus on public API surface first
Missing test files in non-standard locations	Search broadly with `rg --files \	rg -i test`
Assuming no test file = no tests	Some projects co-locate tests; check for `#[test]` or `describe(` blocks in source
Ignoring coverage reports that already exist	Always check for existing reports first (Step 2)

Security

Security scanPassed, 90 of 100Pattern checks on the skill's files.
Model reviewPassed

See the full security report

Quality evaluation

Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

SKILL.md in the repositoryCheckingbrokk-untested-paths/SKILL.md

Popularity

7-day trend score: 0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

Claude Code
Cursor
GitHub Copilot
Windsurf
Codex
Gemini CLI
Cline
Amp
Roo Code
Goose
Continue
Kilo Code
Junie
Trae
OpenHands
Qwen Code
AI Agent
Tabnine Chat
Sourcegraph Cody
Amazon Q
Firebase Genkit
Sweep
Devin
Phind
Mentat
Claude Desktop
Aider
GPT Pilot
Plandex
bolt.new
v0
Replit Agent
Anse
TypingMind
BetterChatGPT
LibreChat
Warp AI
Pieces
Fig AI

Embed badge

[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/brokkai/brokk-skills/brokk-untested-paths/badge)](https://verified-skill.com/skills/brokkai/brokk-skills/brokk-untested-paths)

Something wrong with this skill or its page? Tell us.