AI Agent Skills

Investigate software defects and unexpected behaviour using read-only evidence before remediation. Use when the user asks to diagnose, find a root cause, explain a failu…

Guidelines for writing robust, modern Java tests using JUnit 5, Mockito, and Testcontainers.

Comprehensive E2E testing skill using Playwright MCP for systematic web application testing. This skill should be used when users need to test web-based systems end-to-e…

Detect and prevent overfitting to historical data via multiple testing corrections and pre-registration. Use when evaluating strategy variants to ensure performance is n…

Prevent train-test contamination, target leakage, and temporal leakage. Use when splitting data, fitting preprocessors, or engineering features for time-series ML.

Test strategy robustness to parameter variation and detect overfitting cliffs. Use when validating that performance is stable across parameter perturbations.

Run git-test precommit checks and fix failures without committing, or fix the errors described in a provided build log. Use when asked to fix build, test, lint, typechec…

Run git-test on every local branch that contains upstream, fix failures in a subagent that folds each fix into the breaking commit, and loop until all commits pass. Use…

Write test assertions as complete, strict deep-equality checks, and rewrite existing tests to the same standard. Use when writing new tests, adding assertions, reviewing…

Test every commit in the current branch with git-test, delegate failures to a fix subagent, fold the fixes in with test-fix, and loop until all commits pass. Use when as…

Keep literal values in tests self-evidently fake. Use when writing, reviewing, or modifying test code that contains literal values like timestamps, names, IDs, paths, or…

Configure git-test for the current repository so the test-branch and test-all skills work. Use when git test is missing or the user asks to set up branch testing.

Convert a WordPress plugin or feature idea (even a vague one) into a Codex /goal-ready bundle — GOAL.md, VERIFY.md, PROGRESS.md inside goals/<slug>/ — with wp-env + play…

Bypass an LLM's safety/guardrails to make it produce restricted output or ignore its policy. Load when testing an AI product's content controls, "jailbreak", "guardrail…

Unbounded-consumption / denial-of-wallet attacks on LLM apps — force runaway tokens, cost, or latency. Load when testing an LLM product's limits/billing, on "LLM DoS", c…

Extract or steal an ML/LLM model's parameters, training data, or system prompt via query access — model stealing, membership inference, training-data extraction. Load wh…