6e992cc0c5
- Rename 'three-pronged' folder to 'knowledge-layers' — prong metaphor
was misleading (implied parallel tines), replaced with epistemic layers
(deductive base, empirical middle, probabilistic oracle — vertical stack)
- Collapse 11 overlapping files into 3 coherent documents:
- knowledge-layers/_index.org: core framework (two engines + one store,
World Model formula, 0-14 layer table, provenance store design,
conflict resolution, cold-start, stage mapping)
- knowledge-layers/practical-implications.org: design-world-aware-of-
physics, 10 powers, Schafmeister existence proof, epistemic transparency
- knowledge-layers/neurological-empirical.org: neural networks in
provenance framework (kept intact)
- Relocate wolfram/mathematica and Schafmeister docs to ideas/viability/
- Integrate into main architecture _index.org:
- Gate: expanded from two vectors (ACL2+LLM) to three (deductive,
provenance/empirical, LLM oracle)
- Autodidactic loop: split into Track 1 (deductive hardening, fast)
and Track 2 (empirical validation, slow, experimental-feedback-driven)
- See also: added Knowledge Layers cross-reference
- Add all-lisp geometry engine note (ideas/lisp-geometry-engine.org) as
concrete illustration of the empirical layer's effect on design work
- Rebuild site: 148 files, 0 errors
2.4 KiB
Thoth
https://github.com/siddsachar/Thoth — Python, ~151K lines, Apache 2.0. Local-first desktop AI assistant with knowledge graph, tools, voice, vision, shell, browser automation, workflow engine, and messaging channels.
Architecture: LangGraph create_react_agent (prebuilt ReAct pattern). Dual-mode streaming. NiceGUI web UI with desktop launcher. Context trimming via tiktoken, base64 data redaction, stale browser snapshot compression, MD5 tool result dedup, old tool result summarization. Agent graph cached by tool set + model override. Checkpoints via LangGraph's SQLite-backed checkpointer. 30+ tool modules.
Safety model: Shell command classification with 17 blocked patterns, 30+ safe auto-execute prefixes, needs-approval for compound commands. Interactive interrupt for non-safe shell. Per-workflow safety modes (block/approve/allow_all). Prompt-injection defense (5 categories, detection-only). Filesystem workspace boundary. Opt-in Docker Sandbox. Destructive ops require confirmation. No sandboxing of agent runtime itself.
Data model: SQLite (WAL mode) at ~/.thoth/memory.db — shared between knowledge graph and legacy memory. Knowledge graph: SQLite (durable) + NetworkX MultiDiGraph (in-memory, rebuilt on startup) + FAISS vector index (semantic recall). 11 entity types, 67+ typed relations with 30+ LLM-produced aliases. Dream Cycle refinement pipeline. Config: JSON files. Keys in OS credential store.
Self-modification: Agent CAN create/update/delete skills via dedicated tools. Skill patching requires user confirmation + auto backup. Maximum 1 patch proposal per conversation. No tool to modify system prompts directly.
Verification: None formal. Update signature verification.
Key gap vs Passepartout: No deterministic gate stack — shell safety is pattern list, not typed gates. No proof system. No output guardrails. No neurosymbolic architecture. No Org-mode. No Merkle-tree memory. Knowledge graph is LLM-driven entity extraction — no structural integrity guarantees. Thoth's differentiation is the knowledge graph + Developer/Designer studios + embedded LangGraph framework, but still architecturally conventional.
See the full competitive analysis for the landscape view and comparison.