View or change the Live Memory plugin's configuration for this workspace — the provider profile the memory LLM uses, retention caps, the external-edit watch glob, the compaction interval, and the context-window budget/threshold. Use when the user asks to…
Provides context about the Shofer evals system structure in this monorepo. Use when tasks mention "evals", "evaluation", "eval runs", "eval exercises", or working with the evals infrastructure. Helps distinguish between the evals execution system…
Show what the Second Brain has observed, spent and said — observed volume per segment, pass latency, window fill, token use and cost, advisories generated / gated / delivered, and uptake per detector. Use when the user asks whether the Second Brain is…
Drop the Second Brain's task ledger — one task's, or all of them. A ledger is derived state (goal, compaction notes, advisories, suppressions), so deleting one is always safe; a running task rebuilds from its next observations.
Ask the Second Brain to look now — runs one observer pass immediately over the focused task and reports each detector's verdict plus any advisory. Bypasses the pass throttle, not the mute or the budget. The fastest way to see it work.
Show the Second Brain's recent advisories with their evidence and adjudicated verdicts, plus the ones the gate dropped and why. Use when the user asks what the Second Brain said, why it said it, or why it stayed silent.
Empty the Live Memory for this workspace — wipe its accumulated observation/Q&A log, its persisted conversation window, and its cost ledger so the next question starts from a blank slate. Use when the user wants to reset or clear Live Memory, discard what it…
Show the Live Memory plugin's current status for this workspace — agent state, context-window fill, retained observations and Q&A pairs, session token cost, and the pending question queue. Use when the user asks how Live Memory is doing, how full its context…