web-cliplibrary
Run the following command in DeepSeek Harness:
dsh plugin install ChenneyZhuang/web-cliplibrary
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install ChenneyZhuang/web-cliplibrary in your terminal to install this plugin; the full source is at https://github.com/ChenneyZhuang/web-cliplibrary
About this plugin
An agent tasked with research will plough through a dozen pages in a single session, yet the facts that actually mattered end up as a paraphrase buried in chat scrollback. Three days later you cannot say which page said what — the site may have redesigned, the URL may be dead. Research that lives only in conversation history is, for all practical purposes, lost.
web-cliplibrary fixes the chain with a simple rule: clip at read time. Every clip is a verbatim excerpt paired with its source URL and the date it was fetched, written to a plain-text clips/.md file per project. New entries go on top so the file stays greppable; anything the agent has rephrased is explicitly flagged as a note (mine) and kept apart from the quote. When a deliverable cites a clip, the URL and fetch date travel with it, giving readers a one-link path back to a frozen snapshot of the original page.
The discipline that makes the library trustworthy is strict: capture the moment you are reading, because that page will not be the same page tomorrow. Even a 403 or a bot-wall gets a clip — that error is evidence about the source and saves the next session from blindly re-hitting the same wall. If your daily work involves agent-driven literature review, report writing, or any task where every claim needs a traceable web origin, this is a lightweight layer worth pulling into your workflow.
Use Cases
- After an agent sweeps through a dozen pages in one research run, trace every claim back to its exact URL and fetch date
- Keep a persistent record across multi-turn sessions so that a redesigned or dead page does not break the citation chain
- Attach the source link and retrieval date to every quote in a deliverable so readers can verify the frozen snapshot in one click
Best For
- Researchers who rely on agents for multi-page web research and information synthesis
- Technical writers producing reports or papers where every citation must be traceable to a live source
- Developers orchestrating multi-turn agent workflows who care about persisting intermediate findings to disk
Related Plugins
Memory layer for coding agents that indexes local session history and auto-recalls relevant context before edits, commands, and failures, with no manual search needed.
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.