Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add liza-mas/liza --skill white-box-red-testinggit clone --depth 1 https://github.com/liza-mas/lizaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/liza-mas/liza/white-box-red-testing)<a href="https://agentmods.dev/skills/liza-mas/liza/white-box-red-testing"><img src="https://agentmods.dev/badge/skills/liza-mas/liza/white-box-red-testing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/liza-mas/liza/white-box-red-testing"><img src="https://agentmods.dev/badge/skills/liza-mas/liza/white-box-red-testing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00063 | $0.00879 |
| Opus 5 | $0.00032 | $0.00439 |
| Sonnet 5 | $0.00013 | $0.00176 |
| Haiku 4.5 | $0.00006 | $0.00088 |
Grade A, and why
white-box-red-testing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
White-Box Red Testing
Hard Boundaries
- NEVER modify source code. Analyze and classify only.
- NEVER weaken existing tests.
- Unexported/private members: test only if explicitly scoped.
- Output to user-specified directory or language-appropriate default.
When to Use
- You have specific targets: changed files, a commit range, or low-coverage functions
- You need classified findings (confirmed-bug / likely-bug / specification-gap)
- Contracts exist (docstrings, types, tests) and you want to verify code matches them
- After a commit or PR — "did these changes introduce bugs?"
If you don't know where to look or want to explore broadly, use the black-box-red-testing skill instead.
Scope Modes
User chooses scope. Never run unsolicited.
| Mode | Trigger | Targets |
|---|---|---|
| Commits | --commits HEAD~3..HEAD |
Changed/added functions in diff |
| Files | --files <path> or --module <dir> |
All public callables in specified paths |
| Coverage | --coverage-below 70 [--branch] |
Functions below threshold via scripts/discover_targets.py |
Note on coverage targeting: Low coverage indicates under-tested code, not necessarily buggy code. Use as a targeting heuristic to prioritize where to look, not as a bug predictor.
Workflow
1. IDENTIFY TARGETS
- commits: git diff → parse changed functions
- files: AST parse → list public callables
- coverage: run scripts/discover_targets.py
2. GATHER CONTRACT EVIDENCE per target
- docstrings, type annotations, existing passing tests
- function/param names, assertions, call sites, commit messages
- No evidence? → findings become "specification-gap"
3. FORM HYPOTHESES per target as structured one-liners:
[code_path] × [defect_class] → [observable_symptom]
Defect classes to consider:
- boundary inputs (empty, zero, None, unicode, tz-naive)
- state/mutation (shared state, call sequences, input mutation)
- implicit contracts (name promises vs actual behavior)
- error paths (timeouts, missing data, malformed input)
4. GENERATE ADVERSARIAL TESTS — one test per hypothesis
5. SELF-VALIDATE (mandatory)
- Run all generated tests
- Red → candidate finding
- Green → record hypothesis in confidence section of report
- Broken → fix or discard
6. CLASSIFY per references/finding-classification.md
- confirmed-bug | likely-bug | specification-gap
7. APPLY DISTINCTNESS FILTER
Each finding must differ from others in at least one of:
- The code path exercised
- The category of defect found
- The component boundary tested
Shallow variations of the same finding → consolidate into one.
8. OUTPUT
- Test files to output directory (test code only, no classification in tests)
- Summary report to stdout (classification, evidence, impact per finding)
- Format: see references/output-format.md
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 96 lines · 63 tokens per session scan A f7a6234ed2da
white-box-red-testing is a skill published in the GitHub repository liza-mas/liza (384 stars, last pushed 2d ago), licensed Apache-2.0. It adds 63 tokens to every session and 879 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
engram-testing-coverage
TDD and coverage standards for Engram. Trigger: When implementing behavior changes in any package.
iterative-development
TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically.
python
Python development with ruff, mypy, pytest - TDD and type safety.
nw-fp-clojure
Clojure language-specific patterns, data-first modeling, REPL-driven development, and spec.
strict-tdd
Strict RED->GREEN->REFACTOR test-driven development with enforcement. Never write production code before a failing test. Atomic commits per TDD cycle.
mobiai-mobile-tdd
You MUST use this before writing any implementation code for a mobile feature, bug fix, refactor, or behavior change. Tests come before implementation — no exceptions.