by 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 and a manifest-sealed directory.
# Add to your Claude Code skills
git clone https://github.com/PerryLink/dsh-research-reportGuides for using ai agents skills like dsh-research-report.
Last scanned: 10/5/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-10-05T11:12:59.234Z",
"npmAuditRan": false,
"pipAuditRan": true,
"promptInjectionRan": true
}See how dsh-research-report compares with popular alternatives.
dsh-research-report is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by 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 and a manifest-sealed directory. It has 219 GitHub stars.
Yes. dsh-research-report passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/PerryLink/dsh-research-report" and add it to your Claude Code skills directory (see the Installation section above).
dsh-research-report is primarily written in TypeScript. It is open-source under PerryLink on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh dsh-research-report against similar tools.
No comments yet. Be the first to share your thoughts!
From $49 · 30 days · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM lists this project for discovery and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Automated security scan — not a certification.
交出一份报告,每条结论都能被任何人在自己机器上离线复核——改过任何一个字节,密封哈希就会对不上。
面向 DeepSeek Harness 的可核查研究报告引擎:claim 绑定不可变证据快照,数字与引文按字节核对,结果封存为版本化报告目录,封印就是 dsh-v0.2.2-rc.4(2026-11-18);manifest.json 的 SHA-256。
安装 · 配置 · FAQ · 简体中文 · English · Español · Português · हिन्दी
dsh plugin --profile <name> add dsh-research-report
社区插件,与 DeepSeek 官方无关。唯一仓库:https://github.com/PerryLink/dsh-research-report(Gitee 为镜像)。
给模型一个主题和一组来源,dsh-research-report 把每条结论(claim)绑定到不可变的证据快照、逐字节核对数字与引文,再把结果封存成一份可独立复核的版本化报告。报告目录里有 report.md、manifest.json 和审计日志;封印哈希就是 manifest.json 的 SHA-256,拿到目录的人用 dsh-research-verify 就能离线重算,不需要挂载插件。它刻意不做深研循环:检索与抓取走 ctx.web,长任务走 ctx.jobs,规划与综合留给模型或上游插件。
安装(三选一):
# npm 通道(已发布版本)
dsh plugin --profile web add dsh-research-report
# git 通道(最新 main;等价于 git+https://github.com/PerryLink/dsh-research-report.git)
dsh plugin --profile web add github:PerryLink/dsh-research-report
# tarball 通道:本仓库 pnpm pack 后
dsh plugin --profile web add ./dsh-research-report-<version>.tgz
两条在线通道都会把 bundle 行(见 cordis.patch.yml)装入 profile 的 dsh.profile.bundles 层栈,重启生效。
最小配置(默认即可用;相对路径相对 harness 工作目录解析):
- insert:
- id: dsh-research-report
name: dsh-research-report
config:
ledgerRoot: .research-ledger
reportRoot: research-reports
验证已生效:
dsh --profile web --dump-config | grep dsh-research-report
然后在会话里跑一遍最小闭环,再用 CLI 复核封印:
evidence_add({ origin: "docs/market.md", title: "市场快照" }) # → ev-1a2b3c4d5e6f
research_report({ topic: "...", sections: [...], claims: [...], evidenceRefs: ["ev-1a2b…"] })
ledger_query({ claimId: "c1" }) # 绑定关系 + 核查结论
dsh-research-verify --report research-reports/<topic>/<YYYYMMDD-HHmmss>
卸载:dsh plugin --profile web remove dsh-research-report(或从 profile patch 删掉该行)。
所有 PerryLink 插件均可在 DSH Desktop 内置市场中浏览:市场 → 来源 → 添加来源 → 粘贴 https://perrylink-dsh-catalog.perrylink.workers.dev/catalog-source.json → 选中。安装仍需通过市场的 npm 身份校验与你的确认。
dsh-research-report 的可信度来自三件事:证据不可变、核查可重算、缺口不隐藏。
<ledgerRoot>/objects/<sha256> + JSONL 日志)。同一内容只存一份;快照不可变;每次读取都重算哈希——篡改或删除会被发现,而不是被信任。unverified;证据存在但不足以证实或证伪声称字面量 → insufficient;标签在快照中对应不同数值(且声称值不存在)→ disproven;快照被篡改/缺失 → contradicted。不做语义理解、不做向量相似——只做可审计的字节核对。dsh-data-quality 已挂载时,经其冻结的 verifyCitations 契约做容差核对;数据集不一致即证伪该 claim。10.xxxx/xxxx 结构 + 前缀白名单 + DOI 字符集);非法 DOI 响亮失败。可选期刊/年份元数据被接受;requireJournalMetadata 仅在启用时门禁学术 DOI 证据。<reportRoot>/<slug(topic)>/<YYYYMMDD-HHmmss>/report.md + manifest.json + verification.jsonl + disconfirmation.jsonl;封印哈希 = manifest 的 SHA-256,manifest 内含报告哈希、全部证据哈希与各审计日志哈希。verification.jsonl;verdict drift、绑定证据被篡改/缺失、或审计日志序列化失败都会拦截封存(响亮失败,无 tunable)。disconfirmation.jsonl(claim + 证据引用 + 原因),并在报告的「证伪记录」附录列明。disproofs.jsonl;同一文本在证据未变时被复报会被强制置回 disproven,证据变化后才允许重新核验。verifySealedReport(零网络零模型):重算封印与审计哈希、逐 claim 复核,把 machine-check 段落写入 verifier-note.md;挂载 ctx.jobs 时再派生只读 verifier job(模型复核为增强,绝不替代)。evidence_add 支持可选 sessionRef(sessionId + eventRange,格式校验响亮失败);锚点入账并在附录 B、manifest、verification.jsonl 登记。会话锚定证据诚实判 unverified(会话锚定证据需人工回查会话日志)。[未核实] / [证据不足] / [与证据矛盾] / [已证伪],并在附录 A 列明。绝不静默通过。ctx.web,长任务走 ctx.jobs。规划与综合交给模型(或上游插件)。evidence_add({ origin, content?, title? })——登记一条证据。给 content 则按原文存储;不给时 URL 源经 ctx.web 抓取、workspace 相对路径从磁盘读取(读取绝不逃逸 workspace)。返回证据 id 与 SHA-256 哈希。research_report({ topic, title?, sections, claims, evidenceRefs, gather?, depth?, background? })——组装并封存报告:校验(claim 引用不全响亮拒绝)→ 逐条核查 → 渲染带醒目标记的 report.md → 写 manifest.json → 返回封印哈希。gather: true 先经 ctx.web 做一轮检索,返回已捕获候选证据与显式缺口清单——绝不自动成稿。background: true 经 ctx.jobs 返回 { kind: 'background', jobId }。research_trail({ limit?, format? })——只读的 claim → 证据 → 结论轨迹:把已注册 claim 与证据、已存结论拼起来,把没有可解析证据的 claim 标为 unbound,列出悬空 evidence id,输出可交给审稿人的 Markdown(以及结构化行)。不新增存储、不改写结论。ledger_query({ claimId? | evidenceId? })——只读查询绑定关系与核查结论;证据读取时重算哈希,篡改/丢失会显式报告。不带 id 时返回账本摘要。ctx.researchReport.assemble(request)——面向兄弟插件的冻结服务表面(见 src/service.ts;由 scripts/verify-frozen-contract.mjs 逐字节门禁)。以下为插件真实入口;安装后调用一个即可。
详细示例
命令
research_report({
topic: "示例行业概览",
sections: [{ heading: "市场规模", paragraphs: [{ text: "2025 年全球示例行业市场规模为 1,280 亿元。", claimIds: ["c1"] }] }],
claims: [{ id: "c1", text: "市场规模为 1,280 亿元", evidenceIds: ["ev-5dc0a84d0e48"] }],
evidenceRefs: ["ev-5dc0a84d0e48", "ev-c6cf024e0916", "ev-61f8f4533cb7"]
})
输出
report sealed: /tmp/rr-test-4U2mlF/reports/示例行业概览/20261019-124904
seal (sha256 of manifest.json): 5c85c97597096ced281ca191d7abaaf7434f18b397d35ac81c31aa364a3b12f9
claims: 1 verified / 0 unverified / 0 contradicted / 0 insufficient / 0 disproven (of 1); evidence bound: 3
Observed from README.md:112-118,150; test/index.spec.ts:166-177; src/tools/research-report.ts:170-175
失败分支
输入
{"topic":"probe","sections":[{"heading":"Growth","paragraphs":[{"text":"Growth was 42%.","claimIds":["c1"]}]}],"claims":[{"id":"c1","text":"Growth was 42%.","evidenceIds":[]}]}
输出
edge-case (no throw): isError=false; kind=sealed; verdicts[0]={"claimId":"c1","status":"unverified","note":"claim binds no evidence snapshot"}; counts={"verified":0,"unverified":1,"contradicted":0,"insufficient":0,"disproven":0}; evidenceCount=0; report.md marker=Growth was 42%. [未核实]
Observed from plugins/dsh-research-report/src/verify.ts:173-175 (unverified verdict); plugins/dsh-research-report/src/assemble.ts:209-220 (visible mark); plugins/dsh-research-report/README.md:87 (honest gaps); exercised through plugins/dsh-research-report/test/harness.ts
简短示例:research_report · ledger_query · research_trail
独立二进制(打包为 lib/cli.js,零 @deepseek-ai 导入)无需挂载插件即可审计任意密封报告目录:
dsh-research-verify --report <dir> [--seal <sha256>] [--ledger <dir>] [--format json|sarif]
--report <dir>:密封报告目录(manifest.json + report.md + 审计日志)。--seal <sha256>:期望的 seal 哈希,用于与重算的 manifest 哈希比对;缺省则只报告重算值、不比对。--ledger <dir>:证据账本根目录(objects/<sha256> + index.jsonl),用于逐 claim 字节级复检;缺省则如实跳过 claim 复检。--format:json(默认)或 sarif(SARIF 2.1.0)。它重算 seal 哈希(manifest.json 的 SHA-256)、report.md 哈希与审计日志哈希,逐 claim 重跑字节级 + 完整性检查,任一已执行检查失败则以非零码退出。verifySealedReport / buildVerificationReport / renderSarif / renderVerificationJson 亦从包导出供库调用。
全部可调项都是 Schemastery Config 字段;非法值在加载期响亮失败。相对路径根目录相对 harness 工作目录(workspa