__ __  __  __  ____    ___   ______  __  __  _____   ______  ______
   / //_/ / / / / / __ \  /   | /_  __/ / / / / / ___/  / ____/ / ____/
  / ,<   / / / / / /_/ / / /| |  / /   / /_/ /  \__ \  / __/   / /
 / /| | / /_/ / / _, _/ / ___ | / /   / __  /  ___/ / / /___  / /___
/_/ |_| \____/ /_/ |_| /_/  |_|/_/   /_/ /_/  /____/ /_____/  \____/

software, and the instruments that check it

5         6         7         8         9         10        11        12
|---------|---------|---------|---------|---------|---------|---------|
              |        |        |        |        #        |        |
              0        1        2        3        4        5        6

--[ .plan ]-------------------------------------------------------------

security and correctness in cryptographic and systems code, and the
tools that measure it. most of my work starts with a narrow question and
ends with a result someone else can rerun and, where it turns up
something useful, a patch or a report to the project involved.

--[ upstream ]----------------------------------------------------------

[merged] openssl       #33285     RAND(7): a restore from a saved image
                                  is not detected. all active branches
[merged] botan         #6049      fork safety guidance for snapshot and
                                  checkpoint restores
[merged] circl         #704       wycheproof ml-kem vectors: 1,725 cases
[merged] pmd           #6883      cognitive complexity counts java
                                  switch expressions. released in 7.27.0
[tested] botan         #6054      maintainer's vm rollback fix: repeats
                                  at its parent (15/15 pairs), distinct
                                  with it (30/30)
[review] openssl       #33286     reseed provider drbgs when the linux
                                  vmclock generation counter changes
[review] openssl       #33508     RAND(7): a fork that keeps its pid
[review] circl         #705       wycheproof x25519 and x448 vectors
[review] cargo-mutants #655       --stop-on-missed: stop after the first
                                  missed mutant
[report] acvp-server   #468 #470  gaps between what fips 203 and 204
                                  mandate and what the validation
                                  vectors can exercise. root causes and
                                  measurements on #460, #462, #464
[report] criu          #3237      dump fails on aarch64 when a process
                                  maps /dev/vmclock0; traced the cause
[report] rand          #1843      threadrng after a restore; docs
                                  updated in #1844
[report] node          #66880     crypto random functions after a
                                  restore; docs request

--[ instruments ]-------------------------------------------------------

codecaliper          how readable and complex code is, for python, java,
                     and go. every number cites the versioned rule and
                     grammar behind it
                     [docs] [pypi] [doi]
limen                whether a benchmark comparison clears its own noise
                     floor, from repeated runs, with a ci gate
                     [docs] [pypi] [doi] [action]
cupel                which clauses of fips 203 and 204 the acvp
                     validation vectors exercise, neutralising one check
                     at a time
                     [doi]

--[ systems ]-----------------------------------------------------------

gm-recomp            windows gamemaker games in desktop browsers, from
                     their original executables: x86 recompiled to
                     webassembly, windows apis in rust. ships an
                     embeddable player
                     [releases]
fangame-archive      serverless catalog and reviews for i wanna be the
                     guy fangames: 20,000+ games, 156,000 reviews, 8
                     languages, on cloudflare pages, d1, r2, kv
spaghetti-architect  one program description compiled to python,
                     javascript, go, java, and c++, each run to check
                     they match. also generates labelled llm benchmarks
                     [arxiv] [doi]

--[ stack ]-------------------------------------------------------------

domain     cybersecurity, applied cryptography, fips 203/204
language   rust, c, go, python, java, javascript
runtime    webassembly, webgl2, x86, linux, qemu/kvm, firecracker, criu
method     mutation testing, differential testing, static analysis
platform   cloudflare workers, react, sqlite, github actions

--[ 88x31 ]-------------------------------------------------------------

     
     

kurathsec@github:~$