shofer-dev

shofer-dev/claude-code-live-memory

Plugin Claude Code

Plugin marketplace listing 1 plugin: live-memory.

shofer-dev/claude-code-live-memory

Plugin Claude Code

A cheap, persistent codebase memory for Claude Code — so your agent stops re-reading the repo. A separate large-context model learns your codebase passively across sessions and answers via the asklivememory MCP tool instead of re-loading files; read-only, zero-config on a Claude subscription.

shofer-dev/claude-code-live-memory

MCP server Claude CodeCodexCursor +2

MCP server "live-memory", hosted remotely at 127.0.0.1, as configured in shofer-dev/claude-code-live-memory.

shofer-dev/claude-code-live-memory

Instructions file CodexOpenCode

Instructions for shofer-dev/claude-code-live-memory, covering working agreements — live memory, what this is, stack & conventions, test / gate and running / deploying the server.

shofer-dev/claude-code-live-memory

Instructions file

Instructions for shofer-dev/claude-code-live-memory, a project described as: A cheap, always-on model that learns your repo — so your agent stops re-reading it.

shofer-dev/claude-code-live-memory

Command

View or change the Live Memory's model/provider (writes config.json + hot-reloads the server). Human-facing; not visible to the agent.

shofer-dev/claude-code-live-memory

Command

Empty the Live Memory — wipe its accumulated knowledge for this workspace (or all) so the next question starts from a blank slate. Human-facing.

shofer-dev/claude-code-live-memory

Command

Reset the Live Memory for this workspace — drop the accumulated history trail and re-preload the configured preload globs (verbatim docs) fresh from disk. Human-facing.

shofer-dev/claude-code-live-memory

Command

Show the Live Memory's current status for this workspace (context fill, Q&A retained, cost, queue, uptime). Human-facing; not visible to the agent.

shofer-dev/claude-code-live-memory

Hook

Runs after a tool call finishes for Write, Edit, MultiEdit, NotebookEdit and Read tool calls, executing notify.py via python3 (2 commands). From shofer-dev/claude-code-live-memory.

shofer-dev/claude-code-live-memory

Hook

Runs on the FileChanged event, executing notify.py via python3. From shofer-dev/claude-code-live-memory.

shofer-dev/claude-code-live-memory

Skill Claude CodeCodex

When you need to understand this codebase — where something lives, how a component works, what calls what, conventions — ask the Live Memory instead of re-reading files yourself. It is a persistent, cheap, large-context companion that has already accumulated knowledge of this repo across sessions.

shofer-dev/claude-code-second-brain

Plugin Claude Code

Plugin marketplace listing 1 plugin: second-brain.

shofer-dev/claude-code-second-brain

Plugin Claude Code

A cheap background model that watches your Claude Code session over its shoulder and, rarely, drops one short advisory into it — asynchronously, without ever blocking. It reads only the agent's emissions (29% of the transcript), never its tool results, so it is affordable to leave running for the whole session.

shofer-dev/claude-code-second-brain

Instructions file CodexOpenCode

Instructions for shofer-dev/claude-code-second-brain, covering agents.md, read first, invariants and working here.

shofer-dev/claude-code-second-brain

Instructions file

Instructions for shofer-dev/claude-code-second-brain, a project described as: A cheap background model that watches your session over its shoulder and, when — and only when — it sees something worth saying, drops one short advisory into it: asynchronously, without blocking, interrupting, or being asked.

shofer-dev/claude-code-second-brain

Command

View or change any Second Brain threshold, cap or interval, live. Values validate before they are written and running workers pick them up at the next pass boundary. Human-facing; not visible to the agent.

shofer-dev/claude-code-second-brain

Command

Drop the Second Brain's task ledger — this task, this workspace's, or all of them. A ledger is derived state, so deleting one is always safe. Human-facing; not visible to the agent.

shofer-dev/claude-code-second-brain

Command

Ask the Second Brain to look now — catches the spool up from the transcript, runs one pass immediately, and prints each detector's verdict plus any advisory. Bypasses the pass throttle, not the mute or the budget. Human-facing; not visible to the agent.

shofer-dev/claude-code-second-brain

Command

Show what the Second Brain has observed, spent and said — observed vs dropped volume per tool, pass latency, window fill, token use, and uptake per detector. Human-facing; not visible to the agent.

shofer-dev/claude-code-second-brain

Command

Show the Second Brain's recent advisories with their evidence and adjudicated verdicts, plus the ones the gate dropped and why. Human-facing; not visible to the agent.

shofer-dev/claude-code-second-brain

Hook

Runs when a session starts, executing feed.py via python3. From shofer-dev/claude-code-second-brain.

shofer-dev/claude-code-second-brain

Hook

Runs when you submit a prompt, before the agent sees it, executing feed.py and drain.py via python3 (2 commands). From shofer-dev/claude-code-second-brain.

shofer-dev/claude-code-second-brain

Hook

Runs after a tool call finishes, executing feed.py and drain.py via python3 (2 commands). From shofer-dev/claude-code-second-brain.