This paper describes the design and evaluation of a peer-to-peer cache indexing system, which can be used to integrate the resources of locally available web pages into a globally addressable index using a distributed hash table. The salient feature of the indexing system?s design is the efficient dissemination of cache index information using a next-url index which allows cache clients to determine ahead of time whether linked content is also available at a remote cache. In addition, conventional optimizations such as in browser caching and batching of index write requests are also used. These optimizations are evaluated using trace-driven simulation and the results show that these design trade-offs improve cache lookup performance.