verification 20

DeepWatch's capabilities, installable into an existing DeepSeek Harness profile

baro as a DeepSeek Harness subagent provider: delegate a multi-story goal, get parallel execution, independent review and a verified outcome back — with a live run panel in the sidebar

Multi-agent mathematical problem-solving & verification frameworks for DeepSeek Harness — FOUR agent presets in one install: vibe-math-v2 (probability-driven, classic: qs.json + Propos knowledge base + explorer→solver→review/debate verdict), vibe-math-v3

Source-aware native DSH Topics with manual drafts, permissions, visual boards and learning cards

Talk to Excel in DeepSeek Harness: read, edit, clean and report on spreadsheets by conversation. Re-checks every formula after each edit, repairs silent errors, and needs no local Excel for the core work (charts and PDF export use Windows Excel).

Evidence-first conclusion gate for DeepSeek Harness (DSH): 3-tier evidence hierarchy policy, a model tool that verdicts claims (VERIFIED/REFERENCED/UNVERIFIED), durable per-session tool-activity cross-checking (arguments + output anchors, restart-survivin

LLM-as-a-Verifier for DeepSeek Harness: fine-grained reward, Probabilistic Pivot Tournament best-of-N selection, and per-step progress tracking as agent tools.

Zero-dependency static + sandbox smoke detector for DeepSeek Harness (dsh) plugins: package-structure gates (R), cordis contract scans (K), keyless-headless sandbox smoke (D), and ecosystem-listing checks (CC).

Reference verification for DeepSeek Harness (参考文献真实性检查器): the reference_checker model tool checks a paper's bibliography (.pdf/.bib/.tex/.txt/.md or pasted text) against Crossref, OpenAlex and arXiv, reports found/partial/not_found/error per entry, and em

Privacy-minimal heuristic per-turn execution summaries for DeepSeek Harness

DeepSeek Harness plugin that keeps the agent's green claims honest: verification runs become receipts, edits are tracked, and a stale, failed or masked "all tests pass" is turned into a rerun instead of an ending.

Witness — the quality gate for agent-built web apps. Independent browser acceptance testing: CLI, MCP, AI-drafted checklists, visual regression, multi-browser. Agents self-test and pass; real browsers tell the truth.

Rust dev domain plugin for DeepSeek Harness (dsh): verification pyramid fmt→clippy→build+test→miri+audit(+semver) as model-facing tools + graph-layer governance (anti-Goodhart 8 rules incl. #[allow] density / deleted tests, blind-spot via Cargo.toml & pub

Linux kernel dev domain plugin for DeepSeek Harness (dsh): verification pyramid L1-L4 as model-facing tools + graph-layer governance (anti-Goodhart / blind-spot / patch-conflict) + AGENTS.md contract as a system-prompt section. Plain JS, no build step.

Fact-checking skill for DeepSeek Harness: verifies claims against the open web instead of model memory, cross-checks independent sources across domestic and international search ecosystems, and can route three factual judgements (does the evidence support

Jev (TypeSafe System One decision model) for DeepSeek Harness — seven tools: jev_decision (up to 25 typed questions in one parallel call, measured median 266-484 ms), jev_choose (rank 2-10 approaches), jev_batch (one set of up to 5 judgments over 1-20 tex

Eyes for your DeepSeek Harness agent — it opens, sees, and fixes what it builds. Headless-browser preview & self-verification tools (screenshot, console, DOM read, interact) as a dsh plugin.

Browser verification for DSH plugin development in an isolated environment: throwaway DSH_HOME with its own profile, port and browser instance. DSH 插件开发的隔离环境浏览器验证 skill：临时 DSH_HOME + 独立 verify_<随机> profile 双重隔离，一键拉起隔离 dsh web（自动构建/挂载/启动/清理），不污染正在使用的 web p

DSH plugin: dual-model orchestration for code tasks — a high-capability judge writes Task Contracts and reviews evidence, a cheap worker implements in a git worktree, and the plugin verifies with real commands.

DSH 插件：Agent 声称完成前强制验收，产出标准化验证回执（EvidenceReceipt）