Launches
Loading launches…
Loading launches…
Intelligence
Eval rubric scores, drift, and claim-level provenance.
Before any artifact reaches Gate #2 it is scored by an LLM-as-judge rubric and run through deterministic checks. "Quality Signals" is the operator's view onto three things:
Rubric drift: +0.91 across 5 launches (3.25 → 4.16 / 5).
| Launch | Overall rubric |
|---|---|
| acme/launchpad 93da18… | 3.25 / 5 |
| acme/launchpad 563332… | 3.75 / 5 |
| NousResearch/hermes-agent 49a31f… | 3.75 / 5 |
| acme/launchpad 828322… | 4.44 / 5 |
| OrcaQubits/agentic-commerce-skills-plugins 3b1fed… | 4.16 / 5 |
The judge scores each dimension 1–5; the trend above rolls them into one overall score per launch. Open a launch below for its per-dimension averages.
Open a launch to inspect its rubric scores, drift, and provenance. Counts cover the most recent 5 launches.
| Launch | Repository | Stage | Started | Rubric |
|---|---|---|---|---|
3b1fed7f… | OrcaQubits/agentic-commerce-skills-plugins | Done — Completed | View rubric | |
8283226a… | acme/launchpad | Awaiting you — Artifacts pending review | View rubric | |
49a31f1c… | NousResearch/hermes-agent | Done — Completed | View rubric | |
563332ba… | acme/launchpad | Done — Completed | View rubric | |
93da18ef… | acme/launchpad | Done — Completed | View rubric |