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 squall-chua/skills --skill fault-injection-testgit clone --depth 1 https://github.com/squall-chua/skillsWrote 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/squall-chua/skills/fault-injection-test)<a href="https://agentmods.dev/skills/squall-chua/skills/fault-injection-test"><img src="https://agentmods.dev/badge/skills/squall-chua/skills/fault-injection-test/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/squall-chua/skills/fault-injection-test"><img src="https://agentmods.dev/badge/skills/squall-chua/skills/fault-injection-test.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 | $0.00040 | $0.05721 |
| Opus 5 | $0.00020 | $0.02861 |
| Sonnet 5 | $0.00008 | $0.01144 |
| Haiku 4.5 | $0.00004 | $0.00572 |
Grade A, and why
fault-injection-test 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 10d 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 — 456 lines — stays where its author put it; the contents beside it link to each section on GitHub.
A steady state is the handful of numbers that say the system is serving people: the success rate, the latency at the tail, the orders going through per minute. Write them down and measure them before anything breaks. Without that baseline there is nothing to compare a broken run against, and "it seemed fine" is the whole finding.
Then break the environment underneath it — the database refusing connections, a third party answering 429, a pod killed mid-request, the disk full — and watch what the steady state does. The measurement is a comparison, not a score: how far it moved while the fault was live, how far the damage spread, and whether it came back once the fault was gone.
That last half is the one people forget. A system that survives every fault and then never recovers is worse than one that dips and returns, because a fault ends and this does not.
This skill reports
Steps 1 to 8 run the experiments and write up what happened. The production code stays exactly as it is. That report is the whole deliverable.
Steps 9 and 10 run only on a fix signal: "fix it", "close the gaps", "make it survive", "go ahead". The step 8 report is the before, so it is written on a fix run too, and written before anything is changed.
1. Define the steady state
Before naming a single fault. Pick the few signals that say the system is doing its job, and prefer the ones a customer would feel over the ones a dashboard happens to have.
| Signal | Reads like |
|---|---|
| Success rate | non-error responses ÷ requests, on the paths that matter |
| Latency at the tail | p95 and p99, not the mean — the mean hides the fault you are looking for |
| Throughput | requests, jobs, or messages per second |
| A business number | orders placed, payments captured, messages delivered per minute |
| Backlog | queue depth, consumer lag, connection pool waiters |
| Recovery | how long after a fault stops before the above are back in band |
Each needs three things: the command or query that reads it, a value measured now, and a tolerance band — the range that still counts as serving people. A band decided after seeing the broken run is a band fitted to the result.
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.
- 10d ago First seen · 456 lines · 40 tokens per session scan A 9d846bb37e98
fault-injection-test is a skill published in the GitHub repository squall-chua/skills (2 stars, last pushed yesterday), licensed MIT. It adds 40 tokens to every session and 5,721 once invoked, about $0.0002 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.