Dodgeball
Coordinate a bounded volley of sequential and concurrent defensive tests to reproduce race, contention, ordering, retry, load, or multi-input failures. Use only when the user explicitly invokes Dodgeball for an authorized non-production target or approved safe test window.From its SKILL.md
npx -y skills add jawsublime-byte/simon-says --skill dodgeballAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.3 KB, 183 tokens by cl100k_base, as published. Nobody here has run it
Dodgeball
Throw controlled tests from several angles while keeping every throw attributable.
Safety gate
Lock the target, environment, concurrency ceiling, data fixtures, stop conditions, and prohibited effects. Default to a sandbox. Stop on data corruption, uncontrolled resource growth, unexpected external traffic, or scope escape.
Volley
- Establish a single-request baseline.
- Run relevant cases sequentially to identify independent failures.
- Combine only the cases needed to test ordering or interaction.
- Increase concurrency in declared steps, never as an unbounded flood.
- Record timing, order, state, identifiers, resources, and outcomes for every volley.
- Reproduce each failure with the smallest number of coordinated throws.
- Convert confirmed failures into deterministic tests.
Completion
Report the safe limits, test matrix, failures, minimized reproductions, and untested combinations. Finish only after cleanup succeeds.
What ships with it: 1 file
231 B alongside SKILL.md
agents/
- openai.yaml231 B