NOTES FROM READING GROUP MEETINGS:
01/28/99An attack on modified Skipjack
02/01/99Another attack on modified Skipjack
SKIPJACK AND KEA RESOURCES:
The Skipjack and KEA algorithms were released by the NSA in two PDF files which were, unfortunately, difficult to print and read. The specifications were converted to a Postscript file by various members of the cryptography community. C code for the algorithm is available via anonymous ftp. Note that this is a site in Finland. Posting this code on a domestic web site would be a violation of U.S. export restrictions.
Biham and Shamir have done the first extensive analysis of the Skipjack algorithm, including an attack on a slightly modified version of the algorithm.
Bruce Schneier has interesting comments on Skipjack and KEA in his monthly newsletter. See the July, 1998, August, 1998 and January, 1999 issues.
The Skipjack and KEA algorithms were introduced as part of a controversial attempt by the government to impose a uniform cryptographic system in the United States, one which would allow the government to decipher encrypted messages without the knowledge or permission of the parties exchanging the message. One small part of the firestorm that erupted was caused by the fact that the government was asking the public to trust a classified algorithm, which was contrary to the firm opinion of cryptanalysts that an algorithm should only be trusted if it was public and had undergone extensive peer review. To answer these concerns, the NSA assembled a panel of outside experts and permitted them to review classified material related to the design and testing of Skipjack. This team issued an unclassified report of their review, which was favorable.
For more information on the "Clipper Chip" controversy, see the CPSR web page.