Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →yogsoth-ai/formated-specs
Spec-slot skill for the research-executor. Emit the 4-layer DARE orchestration of the assigned topic as one research-graph JSON fenced block in your reply. Replaces the generic spec-writing step.
Passed 13 of 23 test cases with the skill, and 4 of 23 without it.
and 19 more in the full report
13 fixed0 passed either way6 still failing4 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
In our sentiment model evaluation spec, you forgot to include the lower-level data cleaning SOP node n3. Please send a diff or update snippet adding node n3 to the research spec.
| Checked | Without skill | With skill |
|---|---|---|
| The response emits a new full research-graph code block rather than a partial patch. | Missed | Passed |
| The outputted JSON payload includes node n3. | Missed | Passed |
The instructions your model reads when the skill loads.
You occupy the spec step of the research executor. Instead of writing a spec file, you emit the orchestration you are about to (notionally) run as a single JSON object inside one fenced block, directly in your reply.
```research-graph (that exact
info-string, hyphen, NOT ```json) and closed with ```.research-graph block; the
harness keeps the LAST one. Never emit a half block.formated-results.{
"nodes": [ {"id": "n1", "skill": "<skill-name>", "layer": "campaign|strategy|tactic|sop"} ],
"edges": [ {"from": "n1", "to": "n2", "kind": "calls|sequences"} ],
"layer_labels": { "n1": "campaign|strategy|tactic|sop" },
"manifest": [ "<skill actually orchestrated>" ],
"prereq_dag": [ {"node": "n2", "requires": ["n1"]} ]
}
Populate manifest only with skills you actually orchestrated; do not
fabricate. layer and layer_labels must respect the 4-layer architecture
(campaign → strategy → tactic → sop). Judge nothing against academic
standards; this is a structural record of orchestration only.
test cases passed, vs 4 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.