Pinned Loading

1. graph-guard graph-guard Public

   Ontology/graph-aware retrieval over a personal knowledge vault: typed KG + multi-hop PageRank behind a guarded-RAG seam, with an RDF/OWL/SHACL/SPARQL + owlrl fidelity layer. Tier A pragmatic core, …

   Python

2. rag-guard rag-guard Public

   Guarded RAG: grounded answers, refuse-when-unsupported, PII redaction + eval harness. Stdlib core, bring-your-own model.

   Python

3. agent-gate agent-gate Public

   MCP server that adds a fail-closed quality gate and hash-chained receipt ledger to any AI agent workflow.

   Python 2

4. agent-cost-attribution agent-cost-attribution Public

   A token-efficiency meter + playbook for agentic coding: measure where your tokens and dollars go per stage, catch silently-degraded runs, and cut cost without losing capability. Stdlib-only.

   Python

5. fleet-mode fleet-mode Public

   A measured doctrine for orchestrating AI agents — and knowing when not to. Runs as a live Claude Code skill.

   Python 1

6. bow bow Public

   Bow — an all-Claude (Opus 4.8) chief-of-staff agent: systems design + an agent-orchestration doctrine (Fleet Mode). A sanitized build case study.

   Python