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 samuelgudi/iknowkungfu --skill adversarial-test-designgit clone --depth 1 https://github.com/samuelgudi/iknowkungfuWrote 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/samuelgudi/iknowkungfu/adversarial-test-design)<a href="https://agentmods.dev/skills/samuelgudi/iknowkungfu/adversarial-test-design"><img src="https://agentmods.dev/badge/skills/samuelgudi/iknowkungfu/adversarial-test-design/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/samuelgudi/iknowkungfu/adversarial-test-design"><img src="https://agentmods.dev/badge/skills/samuelgudi/iknowkungfu/adversarial-test-design.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 | $0.00058 | $0.01354 |
| Opus 5 | $0.00029 | $0.00677 |
| Sonnet 5 | $0.00012 | $0.00271 |
| Haiku 4.5 | $0.00006 | $0.00135 |
Grade A, and why
adversarial-test-design 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
adversarial-test-design
A test's job is to fail when the code is wrong. A test that passes whether or not the code is correct has done nothing — it is theatre that produces a green checkmark. This skill is about writing tests that genuinely try to break the code, so that a passing suite actually means something.
The mindset shift: do not write a test to confirm the code works. Write a test to catch the code being broken. Those produce different tests.
Use this skill when
- You are writing tests for new code or a bug fix.
- An existing suite is green but bugs still ship — a sign the tests are not biting.
- You are reviewing tests and want to judge whether they would actually catch a regression.
Real-red before green
Before a test is worth anything, you must have seen it fail for the right reason.
- Write the test.
- Run it against code that is wrong — either the not-yet-written implementation, or the pre-fix buggy code for a bug-fix test. Watch it fail. Confirm it failed because of the behaviour you are testing, not because of a typo or a missing import.
- Now make it pass.
A test that has only ever been seen green is unverified. You do not know if it can fail at all. The single most common defect in a test suite is tests that cannot fail.
Adversarial inputs
The happy path is the input the author already had in mind — and the author already made that case work. Bugs live in the inputs the author did not think of. Deliberately reach for those:
- Empty and absent — empty string, empty list, empty file, zero, null, missing optional field.
- Boundaries — first, last, one past the end, exactly the limit, exactly one over.
- Malformed — wrong type, wrong shape, truncated, trailing garbage, wrong encoding.
- Large — input big enough to expose an O(n²), a buffer assumption, a recursion limit.
- Hostile text — unicode, emoji, right-to-left, embedded quotes and separators, the characters that break naive parsing and naive escaping.
- Concurrent / repeated — the same operation twice, two operations interleaved, retry after partial failure.
- The input that violates an unstated assumption — for every "this will always be X" the code assumes, write the test where it is not X.
What ships with it
1 file 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 · 77 lines · 58 tokens per session scan A f0afce5c755f
adversarial-test-design is a skill published in the GitHub repository samuelgudi/iknowkungfu (2 stars, last pushed 1mo ago), licensed MIT. It adds 58 tokens to every session and 1,354 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 2026-08-31.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.