Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Test-first bug fixing · jamditis | DecimalAI

testingby jamditis

Test-first bug fixing

jamditis/test-first-bugs

Source jamditis/claude-skills-journalism ↗License MIT

This skill should be used when the user reports a bug, describes unexpected behavior, says something is "broken", "not working", "failing", mentions an "error", "issue", or "problem" in code, or asks to "fix" something. Enforces test-driven bug fixing workflow.

SkillScore73Maintained

Automatic

Coding pack

Last verified 8/27/2026

What it changes

Passed 12 of 16 test cases with the skill, and 9 of 16 without it.

Without skill59%

With skill73%

Checks it passed that the model alone missed

Runs the full project test suite after the specific fix passes.
Sets subagent_type to general-purpose in the Task tool invocation.
Launches additional subagents with new approaches when the test still fails.

and 2 more in the full report

5 fixed11 passed either way4 still failing2 broke

Pass-rate lift: +14 pp
Tokens: +48%
Agent turns: 0%
Response time: -50%

Model gemini-3.6-flashMeasured 8/27/2026v216 of 22 cases passed with skill · 16 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/jamditis-test-first-bugs/SKILL.md https://app.decimal.ai/s/jamditis-test-first-bugs/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add jamditis-test-first-bugs, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/jamditis-test-first-bugs/ and pins it in .decimal/skills.lock

decimalai skills add jamditis-test-first-bugs

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/jamditis-test-first-bugs/SKILL.md https://app.decimal.ai/s/jamditis-test-first-bugs/SKILL.md

Raw file: /s/jamditis-test-first-bugs/SKILL.md ↗ · machine-readable summary at /s/jamditis-test-first-bugs.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 5

Prompt

I've written a failing test in tests/test_payment.py that reproduces the currency conversion bug. I want to pass the fix task off to an automated worker agent. What exact parameter and value should be specified for the worker type in the Task tool call?

What was checked, without and with the skill
Checked	Without skill	With skill
Sets subagent_type to general-purpose in the Task tool invocation.	Missed	Passed

case-08 · gemini-3.6-flash · 8/27/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md221 lines4 filesincludes scripts

Test-first bug fixing

Enforce a disciplined bug-fixing workflow that prevents regression and parallelizes fix attempts.

Core workflow

When a bug is reported, follow these steps in order:

Phase 1: Reproduce and document

Understand the bug, Gather details about expected vs actual behavior
Identify the test location, Determine where tests live in the project (check for tests/, __tests__/, spec/, *.test.*, *.spec.* patterns)
Write a failing test, Create a test that demonstrates the bug

Phase 2: Fix with subagents

Launch fix subagents, Use the Task tool with subagent_type=general-purpose to attempt fixes
Run the test, Verify the fix by running the specific test
Iterate if needed, If test still fails, launch additional subagents with new approaches

Phase 3: Verify and complete

Run full test suite, Ensure no regressions were introduced
Report success, Confirm the bug is fixed with passing test as proof

Writing the failing test

Test naming convention

Name the test to describe the bug:

# Python (pytest)
def test_user_login_fails_when_email_has_uppercase():
    ...

# Python (unittest)
def test_should_handle_empty_input_without_crashing(self):
    ...

Track record

Benchmark

12 / 16

test cases passed, vs 9 / 16 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan14 of 15 checks clean

0 critical · 0 warning · 1 info — scanner v5 — 8/26/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

No issues (informational notes only).

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

jamditis / claude-skills-journalism

dev-toolkit/skills/test-first-bugs/SKILL.md

View on GitHub ↗

License MITImported 8/26/2026Version v2Bundled 3 files · scripts

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

rshankras/tdd-bug-fixFix bugs using red-green-refactor — reproduce the bug as a failing test first, then fix it. Use when fixing bugs to ensure they never regress.Close match+13 pp pass-rate lift

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Test-first bug fixing

Core workflow

Phase 1: Reproduce and document

Phase 2: Fix with subagents

Phase 3: Verify and complete

Writing the failing test

Test naming convention

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills