115 tools found
Project State Governor keeps a project's durable, evidence-backed state so an agent entering a fresh conversation knows what's true now.
Multi-Source Search cross-checks claims across independent search providers and outputs a confidence-scored, offline-validatable evidence ledger.
A discipline for maintaining knowledge bases with one fact per file, one-line descriptions for recall matching, and slug-based linking between related files.
DeepRead turns long-form material into an evidence-first reading report that separates claims, evidence, data, examples, assumptions, counterarguments
A durable, source-grounded coding-agent memory skill that combines a Wiki document graph and a live CodeGraph code index.
Research release: a process-scoped Metal capability shim that speeds llama.cpp inference in macOS VMs on Apple Silicon by 7-16x.
Skill for fact-checking AI answers: splits claims, checks citation fidelity, verifies against primary sources and compares platforms in an evidence report.
Real-time detection, instance, and semantic segmentation in one framework and config flag, with multi-backend export (TensorRT, OpenVINO, CoreML).
Research language where every function signature declares its network/file/clock effects, runtime-enforced, for reviewing AI code.
A skill that researches competitor Meta and Google ads and produces a strategic teardown with hooks, funnels, and counter-plays.
Research design choices and best practices with explicit consent before Claude Code writes implementation code.
Runs autonomous Gemini Deep Research tasks that plan, search, and synthesize cited reports, from start to polling to collection.
