Security

论文诚信审计——对一篇论文（外部 PDF/DOI/arXiv 或本地写作产出）做图像误用、数值异常、逻辑缺口三轨证据审计，产出 Level 1–4 分级、可复核的 audit_report.md。用于投稿前自查、审稿前评估、对 AI 生成内容做"数字有没有来源、图表有没有被动过"级核验。触发：integrity…

Perform an explicitly requested local or offline source security audit when a hosted scanner is unavailable or a portable audit is requested; preserve…

Draft, rewrite, audit, and refine text through a specific writing persona using a multi-pass workflow.

Keep a reviewable decision trail for long or unattended work: TSV log per decision, end audit, cross-model attention flags.

Tailscale REST API v2 (api.tailscale.com/api/v2) — devices, keys, ACL, DNS, webhooks, OAuth scopes.

Audit a benchmark comparison for matched model identity, missing coverage, restricted populations and misleading confidence-interval interpretations.

Audit a numeric claim against locally saved source evidence, including denominator and reproducible calculations, before citing or publishing it.

Generate an interactive bash wizard that walks a human through steps only they can perform.

Audit multiple Codex tasks and visible ChatGPT chats for archive and follow-up actions, or recover interrupted Codex tasks when explicitly requested.

Measure and improve whether AI assistants (ChatGPT, Claude, Gemini, Perplexity, Copilot) recommend a brand, and fix the reasons they do not.

The merge lane's runbook and the one home of the integration protocol: who may move main, the lifecycle (produce -> verify -> accept -> land -> deploy ->…

Run the ShipLoop test harness and audit its retained graph, review, test, product and incremental-change evidence.

通过当前福帮手OAuth能力核对本人画像用途、读取已确认事实或提出待确认草案；用途、来源与事实确认分别校验。

Review code using independent adversarial reviewers that search for concrete bugs, contract mismatches, and maintainability regressions.

Build a skill properly — routed tracks, a fresh-context reviewer gate, progressive disclosure, and evals when outputs are checkable.

Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media.

Audit repos for architectural drift, dead code, and abstraction bloat.

Docker and Docker Compose optimization.

Audit benchmark evaluation sets and prompts for training set contamination, n-gram overlap, and judge prompt skew.

Multi-agent adversarial codebase review and deep audit across functional clusters with cross-boundary contract validation and blast-radius scorecards.

Read tweets, threads, bookmarks, articles, and user timelines on X (formerly Twitter) using an authenticated GraphQL flow.

Điều phối multi-agent, tích hợp channel (Telegram/WhatsApp/WeChat), coding agent, monitoring/dashboard, và plugin auth Cursor cho hệ sinh thái Velclaw.

Operate and maintain the Hermes workforce with `flume`: hire/onboard an employee into a repo, the org chart in `~/.hermes/agents-registry.yaml`, the handbook…

Clone API routers with multi-upstream auto-detection.