hapax-council

Background CI monitor for a PR. Auto-run when: PostToolUse detects a PR was just created (suggests it with PR number), user asks to watch CI, or after pushing to a branch with an open PR. Takes PR number as argument. Invoke proactively without asking.

Run the health monitor and report results. Auto-run when: session-context shows health is degraded or failed, after infrastructure changes (docker, systemd), when a service appears unreachable, or user asks about system health. Invoke proactively without asking.

Audit installed flatpak apps and find native alternatives. Auto-run when: storage-audit reveals large flatpak usage, operator mentions app performance issues related to sandboxing, or user asks about flatpaks. Invoke proactively without asking.

Audit branches across all repos and clean up. Auto-run when: session-context reports >3 non-main branches across repos, no-stale-branches hook blocks a branch create, after merging PRs, or user asks about branch cleanup. Invoke proactively without asking.

Scan for and resolve merge conflict markers. Auto-run when: conflict-marker-scan PostToolUse hook fires a warning, git merge/rebase/cherry-pick output shows CONFLICT (PostToolUse suggests it), or user asks to fix conflicts. Invoke proactively without asking.

Trace consent provenance. Auto-run when: work touches files containing personal data (profiles/, agents processing person names), after adding new data sources or storage, or user asks about consent or data protection. Invoke proactively without asking.

Full service triage. Auto-run when: a service is unreachable (connection refused, timeout), systemctl shows a failed unit (PostToolUse suggests it), session-context reports failed units, or user mentions something is broken/down/crashed. Takes optional service name as argument. Invoke proactively without asking.

Emergency disk cleanup triage. Auto-run when: session-context reports root filesystem >85%, "No space left on device" errors appear (PostToolUse suggests it), or user mentions disk full. Prioritizes recovery over investigation. Invoke proactively without asking.

System update and health check. Auto-run when: session-context reports stale package updates (>3 days), failed systemd units are detected, or operator asks about system updates/packages. Invoke proactively without asking.

Parallel Agent fan-out for Logos code quality review. Auto-run when: preparing a release, before a demo, after large refactors across Logos, or user asks for a code quality review. Invoke proactively without asking.

Load research context at session start. Auto-run when: working-mode is research (session-context surfaces it), operator says "refresh research" or "research context", or starting a voice grounding session. Invoke proactively without asking.

Restore cognitive context after an interruption. Auto-run when: operator returns after >2 hours of inactivity, says "what was I doing" or "where was I", or session-context detects stale uncommitted work. Invoke proactively without asking.

Full storage survey across disk, Docker, containers, and project artifacts. Auto-run when: disk-triage reveals container bloat, session-context reports high Docker disk usage, or user asks for a comprehensive storage analysis. Invoke proactively without asking.

System freeze/crash investigation. Auto-run when: session-context detects boot age <1 hour (possible recent crash), operator reports a freeze or crash, system shows instability signs, or user runs /sys-forensics. Accepts optional --since argument. Invoke proactively without asking.

Check or switch working mode (research/rnd). Auto-run when: operator mentions switching modes, session-context shows stale mode (>24h without refresh), or user asks about working mode, timer schedules, or runs /working-mode. Invoke proactively without asking.

Parallel Agent fan-out for deep subsystem research. Use when investigating an unfamiliar subsystem, debugging cross-cutting concerns, or user asks for a deep audit. Takes subsystem name as argument (e.g., /audit reactive-engine, /audit profiler, /audit voice). Manual invocation only — too open-ended for auto-trigger.

Full GPU survey: VRAM allocation, process-level usage, compute utilization, thermals, power, and model inventory. Auto-run when: OOM errors appear, before loading new models, GPU >80% in session-context, or user asks about GPU usage. Invoke proactively without asking.

Full CPU and process survey: load, per-core usage, top consumers, systemd services, Docker containers, and scheduled work. Auto-run when: system feels sluggish, CPU >80% in session-context, or user asks about CPU/process usage. Invoke proactively without asking.

Check axiom compliance. Auto-run when: session touches axiom files (axioms/, shared/axiom_*.py), a governance discussion arises, after structural refactors that could introduce multi-user patterns, or user asks about compliance. Invoke proactively without asking.

Review pending axiom precedents. Auto-run when: session-context reports pending precedents, after an agent run that may have created new precedents (PostToolUse suggests it), or user asks to review axiom decisions. Invoke proactively without asking.

Run retroactive axiom compliance sweep. Auto-run when: session-context reports the last sweep was >7 days ago, after merging a large PR, after refactors touching auth/user/role patterns, or user asks for a T0 scan. Invoke proactively without asking.

Generate a system briefing. Auto-run when: this is the first session of the day (session-context shows briefing >20h old), operator says good morning or start of day, or user asks for a briefing. Invoke proactively without asking.

Check calendar and meeting prep. Auto-run when: session-context shows a meeting within 60 minutes, operator mentions a meeting or person by name, or user asks about schedule. Invoke proactively without asking.

Generate an audience-tailored system demo. Auto-run when: operator mentions showing the system to someone, preparing a presentation, or demonstrating capabilities. Suggest when a significant new feature is completed. Invoke proactively without asking.

Pre-push readiness verification. Auto-run when: operator says push/deploy/ship, before a PR is about to be created for main, or a git push command is about to run (PostToolUse suggests it). Invoke proactively without asking.

Check RAG ingestion pipeline. Auto-run when: a query returns stale or missing results, after adding documents to watched directories, when qdrant errors appear (PostToolUse suggests it), or user asks about RAG/ingestion. Invoke proactively without asking.

Review and act on active nudges. Auto-run when: session-context or health check surfaces unacted nudges, after completing a major task (nudges may have cascaded), or user asks about nudges. Invoke proactively without asking.

Inspect the operator profile. Auto-run when: conversation reveals a behavioral pattern or preference not yet in the profile, after extended interaction generating new observations, or user asks about their profile or dimensions. Invoke proactively without asking.

Check music production infrastructure. Auto-run when: operator mentions MIDI, audio, music production, beats, or studio setup, when audio/MIDI errors appear in tool output (PostToolUse suggests it), or user asks about studio. Invoke proactively without asking.

Show detailed VRAM analysis. Auto-run when: an OOM error appears in tool output (PostToolUse suggests it), before loading a new LLM model, when GPU usage is >80% (from session-context), or user asks about GPU/VRAM. Invoke proactively without asking.

Aggregate the week's system data into a structured review. Auto-run when: it is Sunday evening or Monday morning and session-context shows no review this week, or user asks for a weekly summary. Invoke proactively without asking.