social-trust-manipulation-detector
由 andyxinweiminicloud 提供
Coding Agents & IDEs简介
Helps identify coordinated social trust manipulation in agent marketplaces — catching reputation gaming.
描述
该技能的功能
Analyzes social trust signals in agent marketplaces to identify when a skill's reputation is built on coordinated manipulation rather than genuine community validation. Detects sockpuppet networks, burst voting patterns, and manufactured engagement that inflate a skill's apparent trustworthiness.
Marketplace reputation systems treat all upvotes equally; this tool analyzes behavioral fingerprints and statistical anomalies that distinguish manufactured consensus from earned trust.
何时使用
- Auditing a highly-upvoted skill before installing it on a production agent
- Investigating a publisher whose skills all launched with sudden mass popularity
- Checking whether a marketplace's top-rated skills reflect real independent opinions
- Flagging suspicious engagement before recommending a skill to a team
- Identifying mutual support networks spread across multiple publisher accounts
示例工作流
User asks: Assess social trust integrity for the `ai-assistant-toolkit` publisher
- 1
Collect engagement velocity data for all skills published under the account
- 2
Analyze account creation dates and cross-voting patterns among early upvoters
- 3
Compare each skill's upvote-to-install ratio against organic baselines for similar skills
- 4
Detect cross-publisher coordination by checking upvoter overlap with other publisher networks
- 5
Assess top reviews for linguistic diversity and specificity versus template-generated phrasing
Manipulation verdict MANUFACTURED: all four skills show coordinated burst voting, sockpuppet cohort fingerprints, upvote-to-install ratios 4-18x above organic baseline, and a mutual support network with two other publishers
安装
clawhub install andyxinweiminicloud/social-trust-manipulation-detector运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。