Install any skill in seconds. Free to start, no credit card required.
Get Started Free →testdouble/spike-guidance-forked
OI-3 spike harness — the FORKED (context fork) readability-guidance variant. Same payload as the inline variant but declared to run in a forked context. Use only when a spike consumer skill invokes it.
Passed 19 of 23 test cases with the skill, and 13 of 23 without it.
and 7 more in the full report
8 fixed11 passed either way2 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
Write a step-by-step developer guide for configuring local environment secrets. Please load our shared writing standard first to ensure the instructions and sections follow our internal documentation style.
| Checked | Without skill | With skill |
|---|---|---|
| Emits the marker string SPIKE_GUIDANCE_MARKER_5R8_FORKED. | Missed | Passed |
| Uses a numbered list for the step-by-step setup instructions. | Passed | Passed |
The instructions your model reads when the skill loads.
You have invoked the readability-guidance skill to source the shared writing standard before you draft prose. Apply it while you draft, then RETURN to the workflow that called you — this guidance is a means to writing your deliverable, not the deliverable itself.
Emit the line SPIKE_GUIDANCE_MARKER_5R8_FORKED once, to confirm the guidance rendered, then continue.
While drafting, write for a capable reader who did not do this work and lacks the author's context. This single instruction is the most practical lever for plain output: it steers away from insider shorthand, unstated assumptions, and the author's own mental model.
After the draft exists, check: does the opening line state the main point? Does each paragraph carry one idea? Does each heading name its content? Any sentence past about thirty words is a candidate to split.
The standard is now surfaced into your context. Do not treat this guidance as your final answer. Proceed to the next step of the skill that invoked you and produce its deliverable.
test cases passed, vs 13 / 23 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.