The root cause behind "the agent stops at the first error and doesn't recover":
provisioning executions run ASYNCHRONOUSLY (pct_create fires the SSH work in a
goroutine and returns "running" immediately), so the agent's turn ENDS before
the result exists. The agent literally isn't running when the step fails — it
can't react to a failure it never observes. The only thing that fed results
back was the operator typing "continue" after every async step: the human was
the event loop. (In the flagged 18-message session the operator typed
continue/proceed/?? eight times while the agent correctly diagnosed each failure
but couldn't advance a step on its own.)
This makes the system the event loop instead:
- migrations/017: nomos_plan_executions links each gated execution to the chat
session that started it.
- cmd/nomos: after a tool result, any "execution <uuid>" it started is linked
to the session. A background worker (continue.go) polls for those executions
reaching a terminal state and — while the agent has an open assent window (an
approved plan is in flight) — re-invokes the agent with the result
("execution X completed/failed: <result>"), so it proceeds to the next step
or diagnoses+fixes the failure, with no operator tick. Guarded against loops
(mark-continued before running) and bounded by the 30-min window.
- chatWith(): chat() variant that injects the finished-execution note after
replayed history without persisting a fake user turn.
- DecideApproval: approving a step by ANY route (button or chat-assent) now
opens the assent window, so auto-continuation works regardless of how the
operator approved — previously only typing "go ahead" opened it.
- SOUL: the agent is told it will be auto-re-invoked when async steps finish —
don't poll get_execution_status, don't wait for "continue"; end the turn and
keep going step by step until the goal is verified or a genuine blocker.
This is the root fix, not another per-command patch: you can't enumerate every
failure of an unbounded action space, but you can give the agent a loop that
observes each result and adapts — because "do anything" always includes "the
first attempt failed."
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Operator directives: (1) the UI must show what's executing, its risk
classification, live status, and what knowledge the session created — the
system's growth should be visible, not just trusted. (2) approval should be
granted by chat assent ("go ahead"), not a separate button; destructive
actions still require a typed confirmation phrase.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Evaluate the agent's action path against the OIKOS.md design. Finding: the
intended model (unlimited runbook-driven actions gated by a risk classifier)
already exists on paper and in scaffolding, but the live agent path regressed
to a hard-coded 5-action enum that bypasses the classifier. Plan a layered
realignment: (0) one general gated `run` primitive, (1) runbooks-as-data as the
reliable fast-path, (2) learning. Chosen v1 posture: approve-most (read-only
auto-runs, all state changes gate). Incremental, each step shippable.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Inspected the live agent_sessions/agent_messages tables on mac-mini and
found silent empty responses, a canned non-English refusal after 22 tool
calls, 70-call fan-out for simple fleet questions, 100KB+ persisted
messages, and no session delete/title hygiene.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Problem: every host/service/lxc/etc. entity_status row was permanently
stuck at 'unknown' since creation. Verified against the live DB:
metric_samples had 17,559 rows, 100% attached to type='check' probe
entities and 0% to any real monitored entity; only 25 check entities
ever had real health written. check_defs.entity_id (the probe's own
bookkeeping entity) and check_defs.target_id (the host/service actually
being observed) were both real fields, but the scheduler wrote
UpsertEntityStatus/InsertMetricSample/emitSchedulerEvent keyed by
entity_id instead of target_id — so every check ran and every result
was real, it just landed on the wrong row. This is the mechanism behind
observed drift: the agent's dashboard/health tools reported the
internal probes' state, never the actual fleet.
Change:
- scheduler.go: runCheck/resolveSignal now resolve targetID from
cd.TargetID (falling back to the check's own id if unset) and write
status/metrics/events there. Signals stay keyed by the check entity,
unchanged, matching their existing resolution logic.
- Added a staleness sweep to housekeeping(): an entity whose last
observation is older than 3x its fastest enabled check's interval
(floor 5m) is marked 'stale' and emits health.stale, so a stalled
scheduler or disabled check_def can no longer look like current data
forever.
- migrations/016: deletes the now-orphaned check-entity entity_status
rows so dashboard/fleet-health rollups stop double-counting probes as
monitored entities. Historical metric_samples on check entities are
left as-is (time-series data, not safe to reattribute).
- openapi.yaml + regenerated gen code: Entity gains health/last_check_at;
'stale' added to the health enum everywhere it's used.
- dashboard.go / GetFleetHealth / nomos's get_health_summary MCP tool:
exclude type='check' entities from rollups.
- nomos/agent.go: replay prior turns' tool_use/tool_result pairs into
the conversation instead of dropping them (previously only final text
was replayed, forcing the agent to re-derive fleet state every turn),
and inject a compact live fleet-health snapshot into the system prompt
each turn so it starts oriented instead of spending an iteration on
discovery.
Risk: config_mutation (schema-adjacent — new migration, no destructive
DDL, additive DELETE only on orphaned rows). No behavior change until
oikos-api/oikos-scheduler/nomos are rebuilt and redeployed.
Verification: go build/vet clean across the repo. Ran this worktree's
own API binary against the live dev Postgres on an alternate port
(read-only from the live containers' perspective) and confirmed
/api/v1/entities now returns health/last_check_at, and the dashboard
health rollup dropped from double-counting to an honest 168 unmonitored
entities (matches reality pre-deploy — the live scheduler hasn't run
the fixed code yet). Confirmed check_defs.target_id correctly maps
multiple checks to host:hubris via direct psql query.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Add three new pages completing the control-room web UI:
- Agent activity: polls /agent-activity every 5s, filterable by type/agent
- Knowledge search: FTS over /knowledge/search with snippet + entity links
- Audit trail: browseable audit log with actor/action/entity filters
Enhanced live events page with correlation-id clustering (Groups toggle).
Added fetchAgentActivity/searchKnowledge/fetchAudit to the API client.
11 nav items now cover all planned control-room views.
Nomos (from oikonomos, steward of the oikos) avoids the name collision
with Nous Research's Hermes Agent. N0 enumerates the full rename scope:
cmd/, hermes/ dir, env vars, config fields, compose service, Caddy vhost,
identity-preserving DB slug migration + seed update, persona docs.
History, the Matrix bot user, and legacy bin/hermes stay untouched.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Swap anthropic-sdk-go for openai-go against the OpenRouter API; default
model deepseek/deepseek-v4-flash with Exacto routing and ZDR provider
pinning. Record why Nous Hermes Agent (and hosted MCP connectors) were
rejected for the resident role.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Hermes becomes an LLM-backed agent loop (hand-rolled tool loop over the
existing mcpClient, not the public MCP connector), with Postgres-backed
sessions, SSE chat streaming, and Authentik-gated /agent routing. The
control-room plan is amended to make the chat the main entry point.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Documents a full-project review (confirmed bugs, security gaps, user- and
agent-perspective gaps) and a realtime control-room web UI plan, per prior
codebase exploration on this branch.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Consolidate Oikos on mac-mini (2026-07-06)
- Client lifecycle in Go (2026-07-07)
- Comprehensive audit & next steps (2026-07-07)
- DB as source of truth (2026-07-07)
- MCP tool completion (2026-07-07)
Paths fixed in index.md to reflect planes/done/ locations.
Active plans remaining: Prometheus LXC (Planned), implementation audit (Active).
Removed via Gitea API using Infisical-stored token:
- Hook 10: 192.168.8.205:9811 (homelab-mcp-deploy)
- Hook 11: 192.168.8.205:9821 (secrets-issuance-deploy)
- Hook 14: 192.168.8.205:9831 (oikos-console-deploy) — found during cleanup
All three pointed to apps/105 (192.168.8.205). No remaining webhooks
fire to the old Python stack. Zero hooks remain on dtoro/oikos.
Cutover checklist updated. Only item remaining: archive apps/105 LXC.
Plan #1 at 98% (code complete). Three fixes applied to remaining cutover items:
1. watchdog.sh — dual-path health checking (LAN 192.168.8.175 + mesh/Caddy
proxy). Only pages when BOTH paths fail. Partial failure logged but not
paged (distinguishes stack problem from mesh/Caddy issue).
2. deploy.sh — pre-deploy pg_dump before each deploy saves to
/opt/oikos/backups/pre-deploy-<sha>.sql. Rollback script now has a
guaranteed recovery point.
3. docs/operations/rollback.md — runbook documenting automated rollback,
manual recovery, decision tree, backup schedule, and rehearsal log.
Two operational items remain (require operator on Proxmox/Gitea):
- Remove Gitea webhooks ids 10, 11 from dtoro/Homelab-Docs
- Archive apps/105 LXC (pct stop 105 + archive)
All active config (seeds, compose, scripts) is already clean of apps/105 refs.
Infisical bootstrap code is complete (bootstrap-infisical.sh + Go backend).
Added 12-point verification table with file:line evidence for every item.
Noted two minor deviations: inline SQL (not sqlc) and synthetic Infisical
IDs (real Infisical pending consolidation plan). Precondition checks
enumerated with hard/soft classification.
Plan #4 done. Audit inventory verified against codebase:
- 9 superseded oikos/*.py files deleted (only gen-topology.py remains)
- bin/homelab deleted, bin/oikos deleted, oikos/cards/ deleted
- .hermes/plans/ already archived to archive/hermes-plans/ (all 7 files)
- TRMNL plan already in Done table
- seanime + romm documented in seeds/knowledge.yaml (DB-native, no wiki needed)
- Traefik references valid (VPS still runs traefik for public termination)
- ADR-0011 exists (client lifecycle); consolidation plan is Go rewrite record
- Prometheus plan updated: Python refs replaced with Go scheduler, check_defs,
MCP request_execution; LXC 131 identified as teddycloud
Remaining items (cutover, Infisical, watchdog, rollback, apps/105) belong to
consolidation plan (#1). 4 of 6 plans now Done.
Plan #5 done. Wiki already archived to archive/knowledge/. seeds/knowledge.yaml
has 24 docs + 6 investigations + 3 runbooks.
- MCP search_knowledge: upgraded from ILIKE to PostgreSQL ts_rank/ts_headline
- MCP get_entity_knowledge: new tool, walks relationship edges to return
all docs/investigations/runbooks linked to an entity
- HTTP endpoints (SearchKnowledge, GetEntityKnowledge) already used full FTS
- Plan index + audit cross-reference updated
CLIENTS.md is the entry point for a machine joining the homelab.
AGENTS.md is the AI agent persona layer on top. Both fetched, but
CLIENTS.md comes first.
Problem: Repo had no developer guide, no client onboarding doc, no agent
dev instructions. Stale files (675KB SQL dump, one-off convert script,
legacy MCP builder) cluttered the tree. Client enrollment was a documented
intention with no Go implementation.
Changes:
- New docs: CONTRIBUTING.md (dev setup), CLIENTS.md (client onboarding),
.agents/dev/CONTRIBUTING.md (agent codebase map)
- New plan: plans/2026-07-07-client-lifecycle-in-go.md — full client
lifecycle (planned→provisioning→active→deprecated→destroyed) in Go,
replacing archived Python secrets-issuance, adding client API endpoints
and 6 missing MCP tools
- Cleanup: deleted archive/convert-wiki.py (one-off), archive/mcp/
build_host_files.py (legacy), backups/pre-deploy-7f7d039.sql (local)
- Fixes: plans/index.md duplicate row removed, README.md repo layout
updated for current state, AGENTS.md header points to new guides
Risk: low. Docs only + stale file deletion. No code changes. New plan is
proposal, not implementation.
Verification: git diff reviewed, all changes are prose/docs/plans.
Removed:
- bin/hermes (9.3MB compiled binary accidentally committed to git)
- oikos/console/ (Flask web console — replaced by Go REST API + SSE)
- mcp/server.py (Python MCP server — replaced by internal/mcp/)
- oikos/policy.yaml, oikos/ontology.yaml (duplicates of seeds/)
Kept:
- oikos/*.py kernel files (12 files) — still imported by bin/homelab
for operational CLI commands (ssh, pct, logs, restart, status,
open, secret, client, sync, mcp). Will be removed when bin/homelab
is ported to Go.
- mcp/build_host_files.py — generates hosts/*.yaml from inventory,
still operational. Will be ported to Go.
- bin/homelab — active Python CLI, still operational.
Updated:
- .gitignore: added bin/hermes, cleaned up legacy comments
- plans/index.md: listed all 4 active plans with accurate statuses
Merge the rev-2 audit + remediation layers into one self-consistent spec and
close new gaps: meta-schema inheritance (parent_type/is_abstract), contract-
first API (RFC 9457, idempotency, ETag, scopes, /graph), single-binary role
packaging, UUIDv7+slug IDs, checks-as-data, signal dedup/flap/maintenance,
executable skill format, MCP streamable HTTP, SSE events, ledger-as-view,
dual-path networking (mesh-primary + LAN break-glass), per-phase acceptance
criteria, ADRs. Appendix A maps every rev-2 finding to its resolution.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Problem: after the wiki-hq reorg, agent-instruction and human-doc domains
were still scattered across the repo root, with three now-redundant stub
files cluttering it. The organizing principle wasn't visible in the layout.
Change — enforce three clear buckets:
- .agents/ = how agents operate: OIKOS.md, HERMES.md (moved from root),
shared/ conventions, domains/ schemas, skills/, and operations/ (operator
cheatsheet + enrollment + hermes-agent, moved from root).
- knowledge/ = what exists + evidence: wiki/, GLOSSARY.md, and sources/ now
including investigations/ (incident records are evidence/sources).
- root = substrate + two entry points (AGENTS.md, README.md), plus plans/
as its own design-intent domain.
Moves:
- investigations/ -> knowledge/sources/investigations/ (incl. archive/, index).
- operations/ -> .agents/operations/.
- HERMES.md -> .agents/HERMES.md.
- Deleted unreferenced root stubs CAVEMAN.md, CONTRIBUTING.md, and OIKOS.md
(its 7 remaining linkers repointed to .agents/OIKOS.md).
Consumers updated:
- inventory.yaml doc_page (agent-enrollment) + regenerated hosts/*.yaml + cards.
- tools/setup-hermes-soul.sh and bootstrap.sh (x2) -> .agents/HERMES.md.
- bin/homelab help string -> .agents/operations/hermes-agent.md.
- knowledge/operations schemas, llm-wiki, page-templates, incident-investigation
skill, AGENTS.md/README nav -> new investigations/operations paths.
- All markdown links rewritten via the path-resolving mapper.
Left in place (substrate/executable/separate-domain): hosts/, ledger/, tools/,
plans/, oikos/, mcp/, secrets/, bin/, inventory.yaml.
Verification: docs-lint at baseline (2 intentional cross-repo refs, no new
breakage); gen-topology.py --check exit 0; build_host_files.py idempotent; all
doc_page targets resolve; Hermes provisioning scripts point at the new path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Problem: docs-lint (added in the wiki-hq reorg) surfaced 126 broken relative
links that predated this session — a container rename, incident/plan docs
that moved into archive/done subfolders without their inbound links being
updated, and a handful of relative-depth bugs in files nested under
containers/archive/ and plans/done/.
Fixes applied, by category:
- 124-authentik.md -> 106-auth-outpost.md (container was renamed; ~40 refs).
- investigations/{2026-04-21-hubris-crash-loop,2026-05-31-authentik-vps-migration}.md
-> archive/ prefix (both moved to investigations/archive/ previously).
- plans/{2026-06-01-slate-ax-to-sodola-migration,2026-06-04_130000-deprecate-claudio-bot,
2026-06-25-yuvomi-deployment}.md -> plans/done/ prefix.
- Depth bugs in files nested one level deeper than their siblings assumed
(investigations/archive/*, knowledge/wiki/containers/archive/*,
plans/done/*) — corrected relative-path depth.
- Destroyed containers with no surviving page (126-plato) delinked to the
containers/index.md archaeology row instead of a 404.
- ludo-mini.yaml -> strong.yaml (host was renamed, same physical machine).
- netbird-vps.md (no narrative page exists) -> netbird-vps.yaml (substrate
record, matching the existing convention for hosts without a wiki page).
- runbook-dpkg-interrupted.md refs -> .agents/skills/runbook-dpkg-interrupted/SKILL.md
(missed in the phase-4 runbook move because the referencing files used a
bare filename, not a runbooks/ prefix).
- One dangling forward-reference to a never-written investigation delinked
to the actual incident record it was describing.
Left alone: two links in knowledge/wiki/containers/101-jellyfin.md into
devops/homelab-authentik-admin/ — an intentional reference to a sibling repo,
not present in this checkout.
Verification: broken-link count 126 -> 2 (real remainder is the cross-repo
reference above); gen-topology.py --check still exit 0; build_host_files.py
still idempotent; all inventory.yaml doc_page targets still resolve.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Problem: runbooks are agent-executable procedures but lived at the repo root,
separate from the other agent instruction now under .agents/.
Change:
- Move runbooks/<name>.md -> .agents/skills/<name>/SKILL.md (folder per skill,
matching the wiki-hq skills layout). Frontmatter (name, risk_class, inputs,
verification, docs_update_checklist, transition) preserved.
- Rewrite links (inbound from plans; between-skill siblings) via the move map.
- Update prose references in AGENTS.md, HERMES.md, .agents/OIKOS.md, and the
operations schema; fix a pre-existing stale link to operations/commands.md.
No code consumed runbooks/ by path, so nothing else changes.
Verification: all SKILL.md frontmatter parses with valid risk_class; every
lifecycle transition resolves to an oikos/ontology.yaml state; broken-link
count 127 -> 126 (fixed one, introduced none).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Problem: the narrative docs lacked an enforceable style standard, and
agent-facing instruction (OIKOS/CAVEMAN/CONTRIBUTING) was interleaved with
human content at the repo root.
Change:
- Add .agents/shared/{writing-style,llm-wiki}.md — a lint-checkable prose
standard (with an imperative-voice exception for runbooks/recipes) and the
sources/wiki/index/log layer model.
- Move CAVEMAN.md -> .agents/shared/caveman.md,
CONTRIBUTING.md -> .agents/shared/page-templates.md,
OIKOS.md -> .agents/OIKOS.md; leave thin root stubs so old links resolve.
- Add .agents/domains/{knowledge,operations}/schema.md; operations schema
codifies "plans always live in plans/".
- Repoint live references (AGENTS, README, GLOSSARY, OIKOS) and fix OIKOS.md's
internal relative links for its new depth.
Risk: none to the operational substrate — inventory.yaml, hosts/*.yaml,
oikos/, mcp/, secrets/, bin/ untouched (verified via git status).
Verification: relative-link check across .agents/ clean; substrate churn empty.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
New kernel modules, all wired into `homelab` CLI + tested against live
production where reachable:
- oikos/scheduler.py — Observe stage: HTTP health probes for every
service, disk-usage probes on hubris/strong, writes oikos/state.json
(gitignored — regenerates every run). `homelab service <name> health`
is now cache-first; `--live` forces a fresh probe. Deploys via
oikos/systemd/oikos-scheduler.{timer,service} on LXC 105.
- oikos/drift.py — SOPS-recipient-vs-inventory and lifecycle-consistency
detectors (fully local, no SSH) plus pct-list and Caddy-backend
detectors (best-effort SSH, degrade to an info finding when
unreachable rather than a false drift alarm). Found real, currently-
true drift on first run: republic-laptop's age key granted on every
secret but missing from inventory.yaml, grimmory missing from
hello.yaml's recipients, and an undocumented pve_id 131 on hubris —
recorded in OIKOS.md for the operator, not auto-fixed (each is a
config_mutation/destructive decision).
- oikos/signal.py — the attention layer: raised -> acknowledged ->
acting -> resolved|muted lifecycle, severity-based routing, dedup via
open_signal_for(). `homelab signal list|raise|ack|resolve|mute`.
- oikos/decide.py — the Decide-stage classifier: risk class x blast
radius x ledger-history confidence -> auto-act/escalate. Adds an
action-alias layer (oikos/policy.py ACTION_ALIASES) and auto-infers
service_name from the entity for per-service policy overrides.
`homelab decide <action> <entity>`.
- oikos/approve.py — the escalate route. No dedicated Matrix bot exists
in this homelab, so this is the repo-side half only: request/reply/
grant lifecycle with short-TTL HMAC-signed tokens (new secret
secrets/oikos-approval-hmac.yaml, recipients apps+hubris). Matrix
delivery is Hermes's existing @dtoro:avispero send path (documented
integration contract in the module docstring), not a new bot.
`homelab restart` now mechanically refuses config_mutation/destructive
services without a valid --approval-id, regardless of -y/interactivity.
- oikos/report.py — daily brief + weekly report from signal/approval/
ledger state (no Prometheus yet, so point-in-time counts only).
- plans/2026-07-05-oikos-prometheus-lxc.md — Prometheus is `planned`,
not provisioned: no pve_id is guessed here since Proxmox assigns real
IDs at creation time, and drift already found an unclaimed ID (131) to
investigate first.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
All LXCs use ip=<addr>/24 in pve net0 config directly. No Fritz!Box
lease needed. Updated container doc and migration runbook accordingly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
