A post-change verification workflow that checks whether software actually works after a feature, fix, refactor, or other major change. It includes end-to-end tests, which test a complete user flow, and real-device checks as required gates.
A legal-news briefing skill that drafts updates for an existing client profile, using selected sources and checked cases. It also prepares social-media and public-account article drafts, which remain marked for human review.
A guided preparation step for turning a lawyer’s experience, materials, or case process into a structured brief for building a legal skill. It uses Socratic questions, meaning questions that uncover assumptions through dialogue.
A quality-testing tool for legal skills, where a skill is a reusable instruction set for producing legal work. It combines general skill-lint results with test materials, six evaluation dimensions, scenario-specific checks, and lawyer review.
A case-file-based mock trial organiser that assigns roles such as judge, claimant, defendant, prosecutor, and defence counsel, records their exchanges, and produces a hearing transcript, issue review, and reinforcement checklist.
A local speech-to-text tool that turns audio or video into Markdown transcripts with timestamps and speaker labels. ASR means automatic speech recognition; it runs locally and supports common formats such as MP4, MOV, MP3, WAV, and M4A.
A course-note editor that turns spoken-word transcripts into structured Markdown handouts. It is designed for transcripts from tools such as Tingwu or FunASR.
A version-aware guide for designing, developing, loading, migrating, and publishing plugins for DeepSeek Harness Desktop, a host application for agent extensions.
An unattended overnight workflow that lets a project-management session periodically check worker tasks, review pull requests, merge accepted work, assign queued work, and record where it stopped.
A parcel-tracking and evidence-recording tool for services such as EMS and SF Express. It uses browser automation to check official tracking pages and can archive timestamped PDF proof.
A method for analysing one court judgment. It explains what the court decided, the evidence or rules it relied on, how the reasoning leads to the result, and what may need checking.
A presentation-creation tool for making slide decks from a topic or manuscript and producing editable PPTX or PDF files. It can also study an existing deck's visual style and support structured research for a topic.
A read-only macOS disk-usage analysis workflow. It identifies large or recently growing files such as models, agent sessions, worktrees, dependency caches, and visualisation outputs, and can compare APFS capacity history when available.
A setup tool for configuring how AI models and accounts are used by local coding agents. It can inspect agents, review provider accounts and limits, import accounts, install plugins, configure model routing, and compare task results.
A publishing workflow for turning a private coding-agent skill into something shareable, transferable, or suitable for a public repository. It is intended for authorized publication changes, not ordinary code review or platform syncing.
A tool for scanning local coding-agent conversations, keeping their setup and incremental-processing records, and producing an HTML report of lessons and possible skill improvements.
★not rated 778▲
+10
changed 8d agoA105 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: