Work
Proof of work from Kristopher Richards. Each project is something I built and ran myself.
LLM Red-Team Harness
69 versioned attack cases, deterministic scoring, and hash-chained transcripts. Indirect prompt-injection runs caught canary-token leaks in gemma3:4b and Qwen3-8B.
Read more →Agent memory · Graph Hacks 2026Debias
Memory you can audit: a bias-aware memory graph for AI agents on FalkorDB, with five pure-Cypher detectors and no LLM judge.
Read more →Live range · AI red teamingMarket Leaks
A live prompt-injection range. Five fictional storefronts, five chatbots guarding five vault codes, and one staff door that is definitely not for visitors. Built for the Apart Research Swarm Alignment Sprint, and still standing.
Read more →Applied AI · SMB operationsMargin Watch
A profit-leak finder that shows which small-business orders lost money the moment they were accepted, reconciled to the cent.
Read more →Personal AI · AnalysisChess Coach
1,751 of my own games run through Stockfish into a weakness profile and a conversational coach.
Read more →Shipped game · Logic puzzleLanternfield
A calm logic puzzle about lighting a garden at night. 60 campaign gardens, daily puzzles, free play, and race-a-friend challenge codes.
Read more →Writing · They Can All BirdThe Notebook
My public notebook on intelligence, corvid and machine, home of They Can All Bird and the Convergence Protocol.
Read more →