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 mracal/vibe-engineering --skill adversarial-test-generationgit clone --depth 1 https://github.com/mracal/vibe-engineeringWrote 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/mracal/vibe-engineering/adversarial-test-generation)<a href="https://agentmods.dev/skills/mracal/vibe-engineering/adversarial-test-generation"><img src="https://agentmods.dev/badge/skills/mracal/vibe-engineering/adversarial-test-generation/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/mracal/vibe-engineering/adversarial-test-generation"><img src="https://agentmods.dev/badge/skills/mracal/vibe-engineering/adversarial-test-generation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00044 | $0.01341 |
| Opus 5 | $0.00022 | $0.00671 |
| Sonnet 5 | $0.00009 | $0.00268 |
| Haiku 4.5 | $0.00004 | $0.00134 |
Grade A, and why
vibe-adversarial-test-generation 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 9d 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.
This is a copy
100% identical to vibe-adversarial-test-generation — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 158 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vibe-adversarial-test-generation
Happy-path tests prove your code works. Adversarial tests prove it doesn't break. Spec-linked tests prove it does what it's supposed to.
When to Use This Skill
- After writing happy-path tests for a feature
- Before claiming test coverage is complete
- When spec requirements lack corresponding tests
- When preparing for a security review
- When
vibe-spec-syncorvibe-coverage-enforcerreports uncovered requirements
When NOT to Use This Skill
- Before happy-path tests exist (write those first)
- For throwaway/prototype code
- When the function is trivially simple (e.g., getters/setters)
Modes
Mode 1: Adversarial (Edge Cases)
Generate tests that break assumptions across 6 categories.
1. Boundary Values
- Zero, one, max, max+1 for all numeric inputs
- Empty string, single char, max-length string
- Empty array, single element, very large array
- Exactly at threshold values
2. Nil/Null/Undefined
- nil pointer as receiver
- nil arguments to every parameter
- nil nested fields in structs
- Returning nil where non-nil expected
3. Type Edge Cases
- Unicode: emoji, RTL text, zero-width chars, combining chars
- Strings: newlines, tabs, null bytes, control characters
- Numbers: NaN, Infinity, -0, very large, very small
- Dates: leap year, DST transitions, timezone boundaries, epoch
4. Concurrency
- Two goroutines/threads calling the same function
- Read during write
- Close during use
- Cancel during operation
5. Resource Exhaustion
- Very large inputs (10MB string, 1M element slice)
- Disk full simulation
- Network timeout simulation
- Memory pressure
6. Malformed Input
- Invalid JSON/YAML/XML
- Truncated input (cut off mid-field)
- Wrong types (string where int expected)
- Extra fields, missing required fields
- SQL injection patterns, XSS payloads (for external inputs)
Mode 2: Spec-Driven (Requirement Coverage)
Generate tests that trace back to specific spec requirements.
- Read the spec — Find the specification document for the feature under test
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.
- 9d ago First seen · 158 lines · 44 tokens per session scan A 3873b71ed8f5
vibe-adversarial-test-generation is a skill published in the GitHub repository mracal/vibe-engineering (3 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 1,341 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to vibe-adversarial-test-generation, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
evidence-first-debugging
Use when debugging a crash, lỗi, failing game, tool, build, service, script, or reproducible local code failure requires repro or reproduction, giả thuyết or ranked hypotheses, instrumentation, root-cause isolation, a minimal fix, regression proof, and a regression test.
build-and-runtime-verification
Use when running build, compile, test, launch, or runtime checks and producing a verdict tied to exact commands, exit codes, artifact paths, limitations, and fresh evidence.
load-soak-capacity-verification
Use when servers or online game services need controlled load, soak, concurrency, capacity, saturation, leak, queue, failover, recovery, and scaling verification with baselines and stop conditions; not for client frame budgets.
platform-device-compatibility-matrix
Use when a game needs a platform and device compatibility matrix across OS, hardware, GPU, memory, resolution, input, network, storefront, certification, test evidence, and support policy; not for store metadata preparation.
playtest-evidence
Use when planning or reviewing game playtests that need structured scenarios, participant context, observations, reproduction evidence, severity, and honest separation of observed findings from conclusions.
qa-test-strategy-and-coverage
Use when a game project, milestone, or feature needs a QA test strategy and coverage matrix across risk, test levels, platforms, environments, ownership, automation, regression, evidence, and exit criteria; not for one playtest session.