Score any document. Prove every claim.

Search infrastructure for AI agents: free by default, MCP-ready, LLM-context aware, and production-grade when needed.

Open-source verifier for document citations, source evidence, RAG, and agents, with built-in document parsing.

Trace every claim to its source. Identify every hallucination. Before the report reaches your client.

Pi-native execution contracts for AI agents — evidence-gated completion, ordered plans, atomic decomposition, and compaction-safe resume.

Multi-modal, human-in-the-loop evidence-review agent: one photo → an auditable accept/decline/needs-info verdict. A starter for claims/returns/support agents (CV → Gemini → Claude).

Stage351: Hybrid PQC-ready context-bound enforcement signature manifest layer for Stage350 session_sha256, CI/local context, Sigstore OIDC requirements, and ML-DSA intent metadata.

Mandatory web-first research workflow for any agentic runtime. Three-tier grounding: official docs → research papers → community perspectives. Evidence-verification pipeline with evidence-prediction (not confidence tags). Integrates with findout CLI and skill4eyes. No answer from parametric knowledge alone.

A multi-agent framework for evidence-based verification and safety monitoring of clinical LLM reasoning.

Hybrid Signature Manifest Verification Layer for verification of enforcement session binding, context integrity, hash consistency, Sigstore OIDC requirements, and PQC-ready signature metadata.

Evidence grounded causal reasoning for SemEval-2026 Task 12 (AER) retrieval augmented pipeline with hybrid search, cross-encoder reranking, and calibrated verification. Full research journey from baseline to current version.

Eviblock is a blockchain-based digital evidence verification platform. Upload and verify the authenticity of your files using decentralized trust.

QSP Evidence Match Gate for verifying AI claims against reproduction evidence using target binding, evidence files, SHA256, and signatures.

论文证据驱动的科研工作流 Agent / Paper-grounded research workflow agent for code analysis, safe experiment planning, metric parsing, verification, and reproduction reporting

REMEDA Stage329: submit-ready AI security audit artifacts built on Stage328 evidence match verification.

Stage354: Signature Key Rotation Ledger Layer with Stage178 Assumption, Threat Model, and Guarantee Binding. Transparent key lifecycle tracking, signature key status recording, ledger chaining, and PQC migration-aware verification metadata.

Agentic entity search engine that converts natural language queries into source-grounded entity tables using web search, scraping, LLM extraction, evidence verification, deduplication, ranking, and observability.

Stage339: Unified QSP Behavior Policy Action Gate integrating trust score, evidence match, signature gate, policy reasons, and fail-closed actions.

Stage329 audit submission package extended with automatic SHA256 evidence hash generation for Stage328-compatible reproduction evidence.

Agent-verified work settlement protocol on Somnia.