root
aa55634571
feat: tighten contracts with GPU fleet topology, model routing, Strix Halo details
...
litellm-health: added GPU fleet topology table, model routing map,
3-tier GPU checks (reachability, VRAM, temp, test inference)
litellm-self-heal: added GPU fleet reference, rules for GPU
unreachable and model not responding
Infrastructure topology now fully documented:
- amdpve: Strix Halo CPU (35B, 16 threads, 262K ctx, llama-server)
- llm-gpu: RTX 3090 24GB (Dense tier)
- ocu-llm: RTX 5070 12GB (MoE + Light tiers)
2026-06-29 04:07:05 +00:00
Jerome
9813c12895
v5: Full per-agent isolation separate ledgers, writer registries, canaries. No data crosses agent boundaries.
2026-06-29 01:47:10 +00:00
Jerome
9cf428579c
v4: Add Hermes agent roster (Mumuni, Tanko, Koby, Koonimo), per-agent isolation notes
2026-06-29 01:40:14 +00:00
Jerome
9ac383e812
v3: Add ledger, drift detection, canary checks, per-agent privacy, writer registry
2026-06-29 00:42:54 +00:00
Jerome
e0d98d684a
Improve memory audit contract: autonomous execution, incremental patches, pre-flight checks, stale state thresholds, dry-run mode, success metrics
2026-06-28 19:31:07 +00:00
Jerome
3e10988cce
Improve memory audit contract: autonomous execution, incremental patches, pre-flight checks, stale state thresholds, dry-run mode, success metrics
2026-06-28 19:23:52 +00:00
jerome
94b3bf84b4
Add README + Quickstart sections for agent execution
2026-06-28 13:49:06 -04:00
jerome
6a8a2c61da
Add hermes-config-template prose contract with shared infra + flexible models
2026-06-28 13:33:06 -04:00
jerome
8711840b26
Add memory-audit-maintenance prose contract
2026-06-28 13:20:27 -04:00
root
76fc2595ea
feat(gpu-fleet): GPU fleet management prose contract
...
Declarative GPU roster replaces hardcoded router.py model configs.
Operations: add-model, remove-model, heal, sync-keys, list.
Documents agent key architecture with LITELLM_API_KEY env var.
Covers all 3 GPU hosts (RTX 5070, RTX 3090, Strix Halo).
2026-06-28 16:24:40 +00:00
root
8a190fa803
feat(zulip-health): Gen 3 — stuck detection, proactive queue recovery, restart debounce
...
- Health endpoint now reports stuck:bool + idle_seconds + last_activity_time
- Extension re-registers queue after 30min idle (even without BAD_EVENT_QUEUE_ID)
- Broadened queue expiry detection (matches deregistered, invalid queue, etc.)
- zulip-monitor.sh detects stuck:true + enforces 300s restart debounce
- Prevents silent death where bot shows connected:true but processes 0 messages
Fixes the 'getting stuck and restarts once and for all' issue.
2026-06-28 13:22:11 +00:00
root
90e074dbcc
feat(zulip-health): monitor script + infra report — cron-ready implementations of prose contracts
2026-06-27 19:22:53 +00:00
root
c59afa7f1a
feat(zulip-health): v2 — multi-platform monitor for pi, Hermes, Agent Zero
...
Covers all 4 Zulip agents across 3 platforms with checks, auto-heal,
and consolidated action matrix.
2026-06-27 19:22:44 +00:00
root
14bb145300
feat: expand infrastructure-control pattern — Proxmox 5-node cluster, Docker 3-ecosystem, NFS storage, network services
...
Full discovery and expansion based on PVE API exploration:
Proxmox (5 nodes):
- minipve (.12): authentik, gitea, mumuni, syslog-api, jitsi
- amdpve (.15): abiba, kagentz, tanko, tdunna, baggy, scottdenya
- storepve (.6): docker-vm, ra-h-os, PBS, media, zulip
- acerpve (.9): llm-gpu, adguard
- ocupve (.5): ocu-llm
Docker (3 ecosystems, 22 containers):
- docker-vm (.7): Firecrawl, SearXNG, Home stack, Audiobookshelf
- CT 116: LiteLLM stack (6 containers)
- Netbird (.17): VPN controller + dashboard + crowdsec + traefik
Pattern now includes: access matrix, per-section checks/remediations,
alert routing, escalation chain, quick-start commands, full CT inventory.
2026-06-27 13:11:55 +00:00
root
e090b27e20
feat: expand infrastructure-control pattern — Proxmox, Docker, storage, network
...
Full expansion of the infrastructure-control pattern covering:
Layer 1 — Proxmox Cluster (5 nodes)
- 4 monitoring rules (offline, resource pressure, stopped VMs/CTs, storage)
- Auto-start for 8 critical CTs/VMs on unexpected stop
- Node resource threshold alerting
Layer 2 — Docker Ecosystems (3 hosts, 22 containers)
- 6 monitoring rules across docker-vm, syslog-api, netbird
- Auto-restart for critical containers and stacks
- Disk pressure auto-prune
Layer 3 — Storage Fabric (3.6TB + 7.3TB NFS)
- Mount availability monitoring
- Growth rate tracking for high-use volumes
- PBS backup compliance checking
Layer 4 — Network Services (DNS, auth, git, chat, VPN)
- Service endpoint monitoring
- Escalation to Mumuni for authentik issues
Layer 5 — Agent Health (6 agents across 3 platforms)
- CT auto-start for stopped agents
- Health endpoint verification
Full access matrix and priority-based SLA table.
2026-06-27 13:11:15 +00:00
root
4c6b0410ea
fix: pm2-self-heal — self-process guard, never restart abiba-zulip
...
- Added --no-color flag to all PM2 commands
- Self-process guard: abiba-zulip is READ-ONLY, never restarted
- Corrected column indices (PID=6, restarts=8, status=9)
- Added pm2-self-heal.sh shell script implementation
2026-06-26 19:53:55 +00:00
root
2831b06c1e
feat: zulip-mention-reliability — self-testing contract for @mention detection
...
Monitors whether Abiba Bot responds to @mentions in stream topics.
Self-test: sends a test mention, checks for response within 30s.
Documents 4 known failure modes with remediation rules.
Follows same self-improving pattern as litellm-self-heal.
2026-06-26 19:15:53 +00:00
root
acfe16e5e7
feat: infrastructure-control — reference pattern for environment reach
...
Maps how OpenProse contracts connect to Gitea, containers, remote
agents, and infrastructure. Documents the access matrix, what works
today, and what gaps remain for full enforcement.
2026-06-26 18:52:54 +00:00
root
37de60ae32
feat: add event-driven triggers and success criteria to plugin builder
...
- Success Criteria section defines measurable quality gates
- Continuity section explains event-driven wake modes
- Each generation tracks which criteria passed/failed
- Improvements triggered by failures, not fixed timers
2026-06-26 18:45:06 +00:00
root
be15b9c296
feat: build-zulip-plugin — recursive plugin generation contract
...
Self-improving contract that generates, validates, and iteratively
improves Hermes platform plugins. Tracks generations in knowledge
graph, learns from past successes and failures, and can update
its own rules as new patterns emerge.
2026-06-26 18:39:03 +00:00
root
f340c6301d
feat: pm2-self-heal — monitors and restarts critical PM2 processes
...
Responsibility contract that checks abiba-zulip and abiba-telegram
every 5 minutes. Restarts stopped/errored processes, escalates if
restart count exceeds 5/hour. Logs to knowledge graph and alerts
owner via Zulip DM.
2026-06-26 17:44:27 +00:00
root
41bd3c5748
feat: add daily/weekly digest reporting to self-heal contract
...
- Hourly cycle reports: fix alerts, escalation alerts, 10-cycle all-clear
- Daily digest: 24h summary with counts, top actions, uptime
- Weekly digest: 7-day summary + KG node for trending
- All delivered via Zulip DM to owner
2026-06-26 15:23:21 +00:00
root
fb27c99c3d
feat: litellm-self-heal — proactive remediation responsibility
2026-06-26 14:51:04 +00:00
root
584338fb60
feat: initial OpenProse contracts — LiteLLM health check + hello world
...
- litellm-health.prose.md: Verifies admin UI, API docs, OIDC auth, container
health, and aggregate health endpoint. Reusable by any Syslog agent.
- hello-world.prose.md: Simple example contract demonstrating OpenProse
function pattern with parameters, returns, and postconditions.
2026-06-26 14:40:22 +00:00