Public observation service; Japanese-language records.
Structured observations of ransomware-site listings involving Japan-linked organizations, with source references and observation history.
A listing is a claim, not confirmation of a breach, attribution or data authenticity. The JSON, CSV and feed are published snapshots, not an API for requesting investigations or obtaining leaked data.
Primary sources: Public records Record schema
Agent resources: Observatory machine-readable guide Observation feed
Project Markdown
Released VS Code extension: v0.9.2.
A local VS Code workspace for coordinating tasks across AI providers, following dependencies and reviewing results alongside Git history.
Built-in local task management does not require the Beads CLI; existing Beads workspaces use bd. MIT-licensed software derived from Git Graph, with upstream attribution retained. Provider credentials and local tool permissions remain the operator's responsibility.
Primary sources: Release v0.9.2 Source and license
Agent resources: User guide
Project Markdown
Synthetic-credential demonstrator with recorded cryptographic tests.
An AnonCreds demonstration that checks entitlement and recorded expiry without disclosing names, addresses or licence numbers, and rejects proofs with weakened conditions.
Uses fictional credentials, verifier-configured issuer keys, stored-policy checks and replay rejection. It does not establish a person's identity or current licence status and is not a production identity service.
Primary sources: Source Validation and reproduction
Agent resources: 日本語の技術説明
Project Markdown
Public preprint; related work accepted as an EIML 2026 workshop poster.
Research on stable high-confidence errors in language models and why local robustness need not imply correctness.
The workshop lists Stable Miscalibration in Large Language Models: A Practical View of High-Confidence Errors. The related preprint is titled False Fixed Points: Kantian Feedback, Stable Miscalibration, and Representational Compression in LLMs. Workshop acceptance is not ICML main-conference acceptance or evidence of attendance; the listing does not identify a paper version.
Primary sources: Official workshop listing Related preprint
Project Markdown
