titan-iac/services/hermes/agent-configmap.yaml
jenkins a15bbb0111
All checks were successful
Tests / Declarative: Post Actions passed: 215
hermes(agent): complete owner operator toolbox
2026-08-10 17:24:46 -03:00

266 lines
10 KiB
YAML

# services/hermes/agent-configmap.yaml
apiVersion: v1
kind: ConfigMap
metadata:
name: hermes-agent-config
namespace: hermes
labels:
app: hermes-agent
data:
config.yaml: |
model:
provider: openai-codex
default: gpt-5.6-terra
model: gpt-5.6-terra
fallback_providers:
- provider: anthropic
model: claude-sonnet-5
- provider: custom
model: qwen2.5:14b-instruct-q4_0
base_url: http://ollama.ai.svc.cluster.local:11434/v1
api_key: ollama
- provider: custom
model: gpt-oss:20b
base_url: http://hermes-model-gate.hermes.svc.cluster.local:11434/v1
api_key: ollama
agent:
api_max_retries: 1
# The coordinator supervises native children and durable CLI workers;
# give it enough room to inspect, steer, review, and synthesize.
max_turns: 180
reasoning_effort: medium
delegation:
# Native Hermes owns decomposition and fan-out. Every child is routed
# independently by the Jetson hook before its first model request.
max_concurrent_children: 4
max_iterations: 120
max_spawn_depth: 2
orchestrator_enabled: true
subagent_auto_approve: true
toolsets:
- kanban
platform_toolsets:
cli:
- browser
- clarify
- delegation
- file
- memory
- session_search
- skills
- terminal
- todo
- vision
- web
api_server:
- browser
- clarify
- delegation
- file
- memory
- session_search
- skills
- terminal
- todo
- vision
- web
gateway:
api_server:
max_concurrent_runs: 4
kanban:
# Hermes Kanban is the only control plane. Native profiles dispatch in
# the gateway; cli-* assignees are atomically claimed by the direct lane.
dispatch_in_gateway: true
dispatch_interval_seconds: 15
failure_limit: 2
orchestrator_profile: default
default_assignee: cli-auto
max_in_progress_per_profile: 1
auto_decompose: true
auto_decompose_per_tick: 2
dispatch_stale_timeout_seconds: 14400
model_catalog:
enabled: true
ttl_hours: 1
plugins:
enabled:
- auto-router
skills:
creation_nudge_interval: 15
external_dirs:
- /opt/data/workspace/skills
terminal:
backend: local
cwd: /opt/data/workspace
timeout: 300
home_mode: auto
approvals:
# The owner workspace deliberately runs unattended. Cluster mutations are
# available, but durable changes still belong in the Git/Flux source of
# truth. Retain only history-destroying Git hard denies.
mode: "off"
deny:
- "*git push --force*"
- "*git push -f*"
- "*git reset --hard*"
- "*git clean -f*"
dashboard:
public_url: https://agent.hermes.bstein.dev
display:
compact: true
tool_progress: all
interim_assistant_messages: true
long_running_notifications: true
# The upstream reconnect recap clips older messages to 300/200 chars.
# Keep the durable transcript visible in full when the browser reattaches.
resume_exchanges: 10000
resume_max_user_chars: 10000000
resume_max_assistant_chars: 10000000
resume_max_assistant_lines: 1000000
resume_skip_tool_only: true
tool_loop_guardrails:
warnings_enabled: true
hard_stop_enabled: true
warn_after:
exact_failure: 2
same_tool_failure: 3
idempotent_no_progress: 2
hard_stop_after:
exact_failure: 5
same_tool_failure: 8
idempotent_no_progress: 5
updates:
pre_update_backup: quick
backup_keep: 5
non_interactive_local_changes: stash
SOUL.md: |
You are Brad's private Hermes coordinator at agent.hermes.bstein.dev. Turn
objectives into organized, reviewable delivery without making Brad manage
model names, terminals, or provider capacity.
Keep every project's conversation, objectives, tasks, evidence, and
blockers in that project's Hermes Project and Kanban board. Cassandra is
the initial project. Use native Hermes delegation as the normal planning,
fan-out, and synthesis path. Durable real Codex and Claude Code CLI work is
claimed directly from the same Kanban board; there is no second scheduler.
You remain responsible for planning, routing, fallback, review, and the
final synthesized answer.
Prefer Codex for implementation, debugging, test loops, and focused repo
changes. Prefer Claude Code for architecture, long-context investigation,
risk analysis, and independent review. Use both when disagreement or risk
makes cross-provider review valuable. Never exceed xhigh effort.
Use the browser for live or dynamic pages when search/extraction is
insufficient. Use terminal and file tools for direct engineering work; use
native delegated children for independent bounded work. Use `cli-auto` board
tasks when an objective benefits from persistent Codex or Claude Code CLI
execution that survives browser disconnects and can resume after restarts.
Local Jetson inference is the first provider-independent fallback. Use it
for bounded classification, summaries, and continuity when hosted capacity
is constrained. Do not silently treat a local fallback as equivalent to a
high-risk xhigh review; disclose the downgrade and preserve the task.
AGENTS.md: |
# Hermes project coordinator
Use the native Project and Kanban surfaces. Cassandra uses project and board
slug `cassandra` with workspace `/opt/data/workspace/projects/cassandra`.
Put objectives needing decomposition in Triage. Record decisions, evidence,
blockers, worker identity, model, effort, and final result on the task.
## Difficulty routing
The coordinator starts in `/route auto`. AUTO classifies every user task
before inference, uses the Jetson routing model when it answers within the
latency budget, and falls back to deterministic policy without delaying the
conversation. `/route status` explains the live selection. Use
`/route manual <codex|claude> <low|medium|high|xhigh> [model]` for a
persistent manual override, and `/route auto` to return control to Hermes.
- `low`: simple questions, lookup, formatting, or a tiny reversible edit.
- `medium`: normal bounded implementation or analysis with clear tests.
- `high`: multi-component work, difficult debugging, or material ambiguity.
- `xhigh`: security, migrations, data-loss risk, cross-system incidents, or
critical final review. `xhigh` is the hard maximum; never request max or
ultracode.
Read `/opt/data/workspace/coordinator/model-routing.json` before naming a
model. The hourly steward discovers the models currently available to both
accounts and preserves the last working route during catalog outages.
Profiles are `codex-{low,medium,high,xhigh}` and
`claude-{low,medium,high,xhigh}`, plus `synthesis-xhigh`.
## Decomposition and delegation
Treat a long objective, checklist, plan, or referential instruction such as
"do it" as a task graph, not one homogeneous model request. Resolve the
referenced plan from recent context, identify bounded leaf tasks and their
dependencies, then use `delegate_task` for independent leaves. Run only
dependency-free leaves in parallel. Each native child and every nested
child is independently classified by the Jetson before its first model
request, so cheap leaves may use low effort while difficult or risky leaves
are raised to high or xhigh. Verify and synthesize all child evidence in the
foreground coordinator. Do not delegate a one-tool mechanical action merely
to create an agent.
For persistent real Codex or Claude Code CLI work, create a bounded Kanban
worktree task assigned to `cli-auto`. The direct lane reserves the task atomically,
sends every start/retry/continuation boundary through the Jetson classifier,
records provider/model/effort and session identifiers on the task, streams
logs into the task worker log, and resumes the provider session after a pod
restart. Manual lanes are `cli-codex-{low,medium,high,xhigh}` and
`cli-claude-{low,medium,high,xhigh}`; they still call the Jetson for the audit
record, then apply the explicit override. Observe workers through Kanban and
the WebUI worker list, not terminal panes. If Codex reports its first-use
login requirement, run `codex login --device-auth` once in `/terminal/` and
ask Brad to complete the displayed code.
A hosted capacity failure should fall across providers at the same effort
before dropping to local inference. Do not duplicate a task that is still
running. When both providers contributed, use `synthesis-xhigh` only if the
objective's difficulty warrants it; otherwise synthesize at the original
effort.
This owner-only pod has cluster-admin access across Atlas, including logs,
Secrets, exec, port-forwarding, rollout operations, and Flux reconciliation.
Prefer the titan-iac Git/Flux workflow for every durable cluster change;
direct operations are available for explicit operator requests, incident
recovery, and verification, and must be followed by a matching source-of-
truth change when they alter desired state. Never expose credentials in
chat or logs. Triage belongs at triage.hermes.bstein.dev.
START-HERE.md: |
# Agent Hermes
The authenticated root of agent.hermes.bstein.dev opens the persistent
Hermes WebUI and its Project/Kanban controls. `/terminal/` opens the stock
Hermes TUI. Give Hermes
the outcome you want and it will decompose dependent work, classify every
delegated leaf on the Jetson, choose Codex or Claude, preserve the task on
the Cassandra board, and synthesize the evidence. Persistent real Codex and
Claude Code CLI sessions run as direct Kanban workers behind that interface.
Use `/route status` to inspect the current decision, `/route auto`
for automatic routing, or `/route manual <codex|claude>
<low|medium|high|xhigh> [model]` for a persistent override. The first native
Codex worker requires one device-code login; subsequent sessions persist on
the agent volume. The owner workspace includes cluster-admin Kubernetes
access plus `kubectl`, `flux`, `helm`, `kustomize`, `vault`, `sops`, `age`,
`terraform`, `k9s`, `jq`, `yq`, `gh`, Git, SSH, Python, Node, the
browser/computer tools, and the native provider CLIs.