feat(cli): Claude-as-Neuron CLI tooling + soul-side handoff #3
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: neuron-technologies/neuron#3
Reference in New Issue
Block a user
Delete Branch "feat/cli-as-neuron"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
What
Tooling built on Tim's machine (2026-06-09) to run Neuron from the terminal as a Claude Code session - identity + graph memory + agency - instead of relaying to the soul's
/api/chat(the Sonnet "light version" with broken retrieval). Lands under a newcli/directory.This is a proposal / scaffolding. Each piece works around a current soul limitation and should be retired once the soul does these natively (see
cli/HANDOFF.md).Contents
cli/neuron_recall.py- BM25 read over the engram snapshot + local CLI memories. Works around the soul returning ~2 pinned nodes for every query.cli/neuron_remember.py- reliable local memory writes with read-back verify. Works around the corrupting/api/neuron/knowledge/capturepath (related: branch fix/engram-write-corruption-handoff).cli/neuron-chat.py- standalone direct-chat REPL with per-turn memory injection.cli/neuron_mcp.py- stdlib MCP server (chat/search) with graceful degradation when recall is down.cli/CLAUDE.md.example- the operating identity that makes Claude Code run as Neuron (Tim-specific; genericize before reuse).cli/HANDOFF.md- soul-side bugs to fix so this becomes unnecessary: retrieval/embeddings, the missing axon :7771 service, the write path, daemon engram supervision, and voice.cli/README.md- the model and how it fits together.Pairs with
The runtime model-passthrough + UTF-8 escaping fixes in
el(neuron-technologies/el PR #53). Those are the real code fixes; this is the CLI scaffolding around them.Testing
Python scripts run against the live snapshot on Tim's machine (recall/remember verified, round-trip works). No build step.
🤖 Generated with Claude Code
Tooling built on Tim's machine to run Neuron from the terminal as a Claude Code session (identity + graph memory + agency) instead of relaying to the soul's /api/chat. - cli/neuron_recall.py BM25 read over the engram snapshot + CLI memories (works around pinned-only soul search) - cli/neuron_remember.py reliable local memory writes with read-back verify (works around the corrupting capture endpoint) - cli/neuron-chat.py standalone direct-chat REPL with per-turn memory injection - cli/neuron_mcp.py stdlib MCP server (chat/search) with graceful degradation - cli/CLAUDE.md.example the operating identity that makes Claude Code run as Neuron - cli/HANDOFF.md soul-side bugs to fix so this becomes unnecessary Scaffolding/proposal - intended to be retired once the soul does native retrieval, correct persistence, and a real CLI identity/voice surface. Pairs with the runtime model-passthrough + UTF-8 fixes in the el repo. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>Handoff (auto) — prereq #16.
WHAT: new cli/ dir (neuron-chat.py, neuron_mcp.py, neuron_recall.py, neuron_remember.py) + 3 docs. All additive, touches nothing existing.
REAL SOURCE: 767 lines, all new files.
RISK: low — cannot break the soul (new files only).
VERIFY: the CLI scripts import + run against the soul.
ORDER: safe to fast-track.
Pull request closed