weft-evidence
DiscoveredReproducible evidence for a single-machine multi-agent harness: 20-task A/B regression (0.97-1.00 vs 0.02-0.07), scoring rubric, methodology and limitations, metrics ledger.
Reproducible evidence for a single-machine multi-agent harness: 20-task A/B regression (0.97-1.00 vs 0.02-0.07), scoring rubric, methodology and limitations, metrics ledger.
Comments
Loading…
Similar plugins
by dongsheng123132
Reproducible multi-source release evidence for DeepSeek Harness artifacts and mirrors
★ 2
MIT
JavaScript
Sep 7, 2026
dsh plugin --profile web add dsh-release-proofby dongsheng123132
Deterministic revision-pinned benchmarks and regression evidence for DeepSeek Harness
★ 3
MIT
JavaScript
Oct 5, 2026
dsh plugin --profile web add dsh-benchmarkby PerryLink
Verifiable research-report engine for DeepSeek Harness: content-addressed evidence ledger (claim-snapshot binding, tamper-evident) plus versioned sealed reports with per-claim verification verdicts an
★ 213
↓ 1.1k/wk
Apache-2.0
TypeScript
Oct 18, 2026
dsh plugin --profile agent add dsh-research-reportby AaronandWork
Evidence-first conclusion gate that cross-checks first-hand claims against a durable per-session tool-activity index, blocks unverified guesses and tracks blocked-versus-converted metrics.
★ 5
↓ 204/wk
MIT
JavaScript
Sep 12, 2026
dsh plugin --profile web add @aaronandwork/dsh-evidence-gateby zhengqiuyang
Compaction integrity for DeepSeek Harness — constraints from the compacted range are verified against the summary; what the agent was told to remember and silently forgot is reported loudly
★ 0
MIT
JavaScript
Oct 17, 2026
dsh plugin --profile web add dsh-compaction-guardby Harzva
DSH-native multi-runtime baseline, ablation, and reproducible evaluation control plane
★ 0
MIT
TypeScript
Aug 17, 2026
dsh plugin --profile web add @harness-flow/dsh-research-lab