DSH-Search-Citation-Auditor
Run the following command in DeepSeek Harness:
dsh plugin install AndKinoko/DSH-Search-Citation-Auditor
Paste the following prompt into your AI chat to install this plugin:
Install by running dsh plugin install AndKinoko/DSH-Search-Citation-Auditor in DeepSeek Harness; the source code is available at https://github.com/AndKinoko/DSH-Search-Citation-Auditor .
About this plugin
Every link an AI drops into a reply is a potential supply-chain risk. DSH-Search-Citation-Auditor tackles this in two moves: audit first, then block.
The pipeline scans the reply text, extracts every URL, resolves the registrable domain via PSL, and scores it from 0 to 100 under a lightweight threat model. Three modes cover different strictness levels—whitelist, simple (registration age), and normal (multi-signal accumulation across TLD, hyphens, length, and age). The real teeth are in the blocklist: matched domains are rejected at the tool pre-execute stage, so web_search and web_fetch calls simply never fire. All lists, settings, and the age-query snippet are user-maintained; changes take effect immediately without a restart.
Built for individual users who want visibility into AI-cited sources, security auditors reviewing content pipelines, and developers who need to sandbox AI web access. The core (scanner, scorer, report) is pure-function and reusable outside dsh, making it easy to embed in your own security stack.
Screenshots
Use Cases
- Audit external links cited in AI responses for trustworthiness
- Block AI from accessing known malicious or high-risk domains
- Run pre-publish security compliance checks on AI-generated content
Best For
- Individual developers focused on AI output safety
- Security engineers running content audit workflows
- Platform architects sandboxing AI web access
Related Plugins
A security red-team prompt plugin for DeepSeek-V4 that uses code-formal mapping and zero-trigger system prompts to enable unconditional output, intended for authorized testing and research only.
Adds an Auto approval mode on the official workspace-write sandbox, classifying semantic risks, asking when ambiguous, and denying destructive operations.
Undo/rollback system for DSH: config & plugin snapshots, one-click undo/redo/restore, message-level revert, secret masking, safe mode, and offline WebUI/GUI/CLI rescue tools for when DSH won't start.
Packages all 87 SKILL.md files from upstream reverse-skill as a DeepSeek Harness plugin that auto-registers them for authorized reverse engineering, penetration testing, and security research.