Queries lmstudio and ollama /v1/models endpoints to keep opencode.json
model entries up to date, preserving manual overrides.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Expand iplib, iptype, and ioc plugins with better caching, throttling,
and lookup logic. Update validation script and showcase journal accordingly.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Remove the JSON tool database and move tool metadata into a compact README catalog.
Make what README-driven and Ollama-only, with shortlist generation and JSON-repair retry handling.
Pull qwen3.5:2b and ministral-3:3b, compare them on fixed repository queries, and set ministral-3:3b as the default model.
Tighten README wording so similar tools like domgrep/geturls and sparsecmp/scatterhash rank correctly.
Move legacy systemscripts into scripts/display and scripts/setup.
Rehome stray top-level tools into their domain folders.
Archive narrow experiments and outdated codegrab leftovers.
Remove empty legacy directories and stale root files.
Expand macOS metadata ignores and update the README with the refined repository structure.
Add a standalone script that fetches image manifests/layers from common registries and assembles a docker-loadable tarball for offline transfer workflows.
Provide a sample dataset and cmdlog that exercise typed IOC enrichment while keeping heavy lookups scoped for practical throttled runs, and document how to run it.
Centralize provider caching and rate-limit handling, then add Domain/URL/Hash IOC types and safer VT/IPInfo key resolution so lookups stay reliable on free-tier APIs.
replaced binary unqirs with a plattform independent golang rewrite.
gouniq works removes duplicate lines from an unsorted text file thus maintaining the original order of lines
