Borrowing it
Nothing to install: this file belongs to adrienm7/ergopti. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/adrienm7/ergopti/main/.agents/skills/meta-test/SKILL.mdgit clone --depth 1 https://github.com/adrienm7/ergoptiWrote 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/adrienm7/ergopti/meta-test)<a href="https://agentmods.dev/skills/adrienm7/ergopti/meta-test"><img src="https://agentmods.dev/badge/skills/adrienm7/ergopti/meta-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/adrienm7/ergopti/meta-test"><img src="https://agentmods.dev/badge/skills/adrienm7/ergopti/meta-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 | $0.00025 | $0.01059 |
| Opus 5.5 | $0.00010 | $0.00424 |
| Sonnet 5.5 | $0.00005 | $0.00212 |
| Haiku 4.5 | $0.00003 | $0.00106 |
Grade A, and why
meta-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 15d 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing source-scanning meta-tests
Meta-tests live in static/ergopti_plus/windows/tests/meta/ and assert
structural invariants — "this hotkey is gated by #HotIf", "this constant is
included before that loader" — that no runtime call can check.
Always read source through the helpers
Defined in tests/test_framework.ahk:
| Helper | Returns |
|---|---|
_DriverSourceConcat() |
the whole driver source, concatenated |
_DriverSourceNoComments() |
same, with comments stripped |
_DriverFuncBody(Name) |
one function's body |
_DriverDirConcat(RelDir) |
every file under one directory |
_StripFullLineComments(Src) |
comment-stripped copy of any source string |
Never hardcode a path like "platform/remap/lalt.ahk". A CI ratchet
(tools/test/test-no-pinned-source-reads.cjs) fails the build when the count of
location-pinned reads rises above BASELINE = 20.
Never raise that baseline to make a change pass — convert the test to a helper read instead. Raising it is the ratchet equivalent of weakening a test.
Why this works: each file's content is contiguous inside the concatenation, so
relative order and "nearest preceding #HotIf" assertions stay valid even after
the file is moved or renamed. State that reasoning in a comment — it is not
obvious to the next reader.
The same rule on macOS — and it is automated
Lua tests use helpers.read_driver_source(symbol), never
helpers.driver_root() .. "modules/foo/bar.lua". The twin ratchet is
tools/test/test-no-pinned-source-reads-lua.cjs.
You should not have to think about this: the pre-commit hook blocks a staged
Lua test that introduces a pinned read, and npm run fix:pinned-reads
converts it for you. npm run lint:pinned-reads surveys the whole tree.
The one thing the fixer will not do for you is guess. read_driver_source
returns every production file containing the symbol, concatenated — so a
selector matching two files changes what a position-comparing assertion means,
silently. The fixer therefore proves a candidate declaration is unique across
the production tree before using it, and refuses when none is
(_on_config_changed is assigned in two files — a real case). When it refuses,
make the assertion order-independent first, then convert by hand.
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.
- 15d ago First seen · 109 lines · 25 tokens per session scan A aeed7d81930f
meta-test is a skill published in the GitHub repository adrienm7/ergopti (14 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 1,059 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-14.
Other skills, from other repositories
CI Pipeline Optimizer
Optimize CI test pipelines through intelligent test splitting, parallelization, caching strategies, and selective test execution based on code changes.
data-scientist
!cat skills/shared/protocols/ux-protocol.md 2>/dev/null || true !cat skills/shared/protocols/input-validation.md 2>/dev/null || true !cat skills/shared/protocols/tool-efficiency.md 2>/dev/null || true !cat .production-grade.yaml 2>/dev/null || echo "No config — using defaults".
Database Performance Testing
Database performance testing including query optimization, index analysis, connection pooling, slow query detection, and N+1 query prevention.
Frontend Performance Testing
Frontend performance testing covering bundle size analysis, render performance, animation frame rate, React re-render detection, and lazy loading verification.
template-code-project
Code-centric optimization research exemplar — gradient descent, convergence analysis, automated figures, dashboard, and publication-quality PDF.
vectorbt-expert
VectorBT backtesting expert. Use when user asks to backtest strategies, create entry/exit signals, analyze portfolio performance, optimize parameters, fetch historical data, use VectorBT/vectorbt, compare strategies, position sizing, equity curves, drawdown charts, or trade analysis. Also triggers for openalgo.ta…