Compare commits

..

2 Commits

Author SHA1 Message Date
d1ba93fd38 feat: add plugin URLs pointing to gortium account
Some checks failed
Build Hermes agent / build (pull_request) Has been cancelled
Build ollama (gfx906) / build (pull_request) Has been cancelled
2026-05-24 19:58:42 -04:00
317a5b23af feat: drop fork, use official image + plugin URL build arg 2026-05-24 19:47:08 -04:00
9 changed files with 21 additions and 568 deletions

View File

@@ -28,8 +28,8 @@ services:
hermes:
build:
context: ./hermes
ssh:
- default
args:
HERMES_PLUGIN_URLS: "git+https://code.lazyworkhorse.net/gortium/hermes-piper-plugin.git;git+https://code.lazyworkhorse.net/gortium/hermes-identity-plugin.git"
container_name: hermes
entrypoint: ["/bin/bash", "-c",
"bash /opt/data/hermes-tools/install.sh && bash /usr/local/bin/run-multi-gateways.sh && exec /usr/bin/tini -g -- /opt/hermes/docker/entrypoint.sh \"$@\"",
@@ -72,8 +72,6 @@ services:
networks:
- ai_backend
- ai_net
depends_on:
- honcho
labels:
- "traefik.enable=true"
- "traefik.docker.network=ai_net"
@@ -127,7 +125,6 @@ services:
- "traefik.http.routers.syncthing-https.tls.certresolver=njalla"
- "traefik.http.services.syncthing.loadbalancer.server.port=8384"
ollama:
build:
context: ./ollama
@@ -161,92 +158,6 @@ services:
- "303"
- "26"
# --- Honcho + OpenConcho combiné: API + Web UI nginx/FastAPI ---
honcho:
build:
context: ./honcho
ssh:
- default
container_name: honcho
restart: unless-stopped
environment:
- DB_CONNECTION_URI=postgresql+psycopg://honcho:honcho_pass@honcho-db:5432/honcho
- CACHE_URL=redis://honcho-redis:6379/0
- CACHE_ENABLED=true
- EMBEDDING_VECTOR_DIMENSIONS=1024
- AUTH_USE_AUTH=true
- AUTH_JWT_SECRET=${HONCHO_AUTH_JWT_SECRET}
# Needed by deriver/dream to make LLM calls (api_key_env = "HONCHO_OPENAI_API_KEY" in config.toml)
- HONCHO_OPENAI_API_KEY=${HONCHO_OPENAI_API_KEY}
volumes:
- honcho_data:/app/data
- /mnt/HoardingCow_docker_data/Honcho/config.toml:/app/config.toml:ro
networks:
- ai_backend
- ai_net
labels:
- "traefik.enable=true"
- "traefik.docker.network=ai_net"
# Router for HTTP + redirect to HTTPS
- "traefik.http.routers.honcho-http.rule=Host(`honcho.lazyworkhorse.net`)"
- "traefik.http.routers.honcho-http.entrypoints=web"
- "traefik.http.routers.honcho-http.middlewares=redirect-to-https"
# Router for HTTPS with TLS — protected by Authelia
- "traefik.http.routers.honcho-https.rule=Host(`honcho.lazyworkhorse.net`)"
- "traefik.http.routers.honcho-https.entrypoints=websecure"
- "traefik.http.routers.honcho-https.tls=true"
- "traefik.http.routers.honcho-https.tls.certresolver=njalla"
- "traefik.http.routers.honcho-https.middlewares=hermes-auth"
# Service Loadbalancer (nginx port)
- "traefik.http.services.honcho.loadbalancer.server.port=80"
depends_on:
honcho-db:
condition: service_healthy
honcho-redis:
condition: service_healthy
honcho-db:
image: pgvector/pgvector:pg15
container_name: honcho-db
restart: unless-stopped
ports:
- "127.0.0.1:5432:5432"
command: ["postgres", "-c", "max_connections=200"]
environment:
- POSTGRES_DB=honcho
- POSTGRES_USER=honcho
- POSTGRES_PASSWORD=honcho_pass
- PGDATA=/var/lib/postgresql/data/pgdata
volumes:
- /mnt/HoardingCow_docker_data/Honcho/postgres:/var/lib/postgresql/data
- ./honcho/init-db.sql:/docker-entrypoint-initdb.d/init.sql:ro
networks:
- ai_backend
healthcheck:
test: ["CMD-SHELL", "pg_isready -U honcho -d honcho"]
interval: 5s
timeout: 5s
retries: 5
honcho-redis:
image: redis:8
container_name: honcho-redis
restart: unless-stopped
ports:
- "127.0.0.1:6379:6379"
volumes:
- /mnt/HoardingCow_docker_data/Honcho/redis:/data
networks:
- ai_backend
healthcheck:
test: ["CMD-SHELL", "redis-cli ping"]
interval: 5s
timeout: 5s
retries: 5
networks:
ai_net:
external: true
@@ -255,11 +166,6 @@ networks:
driver: bridge
name: ai_backend
volumes:
honcho_data:
external: true
name: honcho_data
# llama_cpp_devstral:
# image: ghcr.io/ggml-org/llama.cpp:server-rocm
# container_name: llama_cpp_devstral

View File

@@ -1,33 +1,15 @@
# syntax=docker/dockerfile:1
# Hermes Agent -- custom fork build
# Builds on top of official image + overlays our forked source from Gitea.
# Requires Docker BuildKit. Pass SSH agent for git clone:
# Hermes Agent -- official image + custom plugins layered on top.
# No fork needed — customizations are pip-installable plugins from Gitea.
# docker compose build hermes
# Or manually:
# DOCKER_BUILDKIT=1 docker build --ssh default -t hermes-agent:custom .
# DOCKER_BUILDKIT=1 docker build --build-arg HERMES_PLUGIN_URLS="url1 url2" -t hermes-agent:custom .
# ---------- Base: official Hermes image (system deps, npm, uv, Playwright) ----------
FROM nousresearch/hermes-agent:latest
# ---------- Overlay our forked source ----------
# Uses SSH agent forwarding from the build host (no key baked into image).
# --exclude node_modules/.venv keeps the base image's pre-built layers intact.
# Only the Python source, web UI source, and config change.
RUN --mount=type=ssh \
mkdir -p /root/.ssh && \
ssh-keyscan -p 2222 code.lazyworkhorse.net >> /root/.ssh/known_hosts 2>/dev/null && \
cd /tmp && \
GIT_SSH_COMMAND='ssh -p 2222 -o StrictHostKeyChecking=no' \
git clone --depth 1 --branch main \
git@code.lazyworkhorse.net:gortium/hermes-agent.git fork && \
rm -rf fork/node_modules fork/.venv fork/.git && \
cp -a fork/. /opt/hermes/ && \
rm -rf /tmp/fork /root/.ssh/
# ---------- Reinstall Python package (editable) ----------
# Picks up source changes from our fork.
RUN . /opt/hermes/.venv/bin/activate && \
uv pip install --no-cache-dir --no-deps -e /opt/hermes
# ---------- Plugin URLs (semicolon-separated, set via compose.yml build args) ----------
ARG HERMES_PLUGIN_URLS=""
# ---------- Extra system deps ----------
USER root
@@ -44,16 +26,6 @@ RUN apt-get update && \
COPY --chmod=0755 --from=ghcr.io/astral-sh/uv:latest /uv /usr/local/bin/
# ---------- Matrix bridge + extra pip deps ----------
# Previously installed inline at container startup and persisted via volume mount.
# Now baked into the image so the fragile venv volume mount can be removed.
RUN . /opt/hermes/.venv/bin/activate && \
uv pip install --no-cache-dir 'mautrix[encryption]' openai
WORKDIR /opt/hermes
# ---------- Matrix bridge + extra pip deps ----------
# Previously installed inline at container startup and persisted via volume mount.
# Now baked into the image so the fragile venv volume mount can be removed.
RUN . /opt/hermes/.venv/bin/activate && \
uv pip install --no-cache-dir 'mautrix[encryption]' openai
@@ -84,6 +56,19 @@ os.remove(tgz)
print('himalaya v1.2.0 installed')
PYEOF
# ---------- Install custom plugins from URLs ----------
# HERMES_PLUGIN_URLS is a semicolon-separated list of pip-installable
# package URLs (e.g. git+https:// or direct .tar.gz archives from Gitea).
# Each plugin is installed into the Hermes venv.
RUN if [ -n "$HERMES_PLUGIN_URLS" ]; then \
. /opt/hermes/.venv/bin/activate && \
IFS=';' read -ra URLS <<< "$HERMES_PLUGIN_URLS" && \
for url in "${URLS[@]}"; do \
echo "Installing plugin: $url" && \
uv pip install --no-cache-dir "$url"; \
done; \
fi
# ---------- Install multi-gateway launcher ----------
# Launches one gateway process per profile (HERMES_PROFILES env var)
COPY --chmod=0755 run-multi-gateways.sh /usr/local/bin/run-multi-gateways.sh

View File

@@ -1,75 +0,0 @@
# build stage — fetches and builds Honcho from source
FROM python:3.13-slim-bookworm AS honcho-builder
RUN apt-get update && \
apt-get install -y --no-install-recommends git openssh-client && \
rm -rf /var/lib/apt/lists/*
COPY --from=ghcr.io/astral-sh/uv:0.9.24 /uv /bin/uv
ARG HONCHO_REPO=ssh://git@code.lazyworkhorse.net:2222/Hermes/honcho.git
ARG HONCHO_REF=main
RUN mkdir -p -m 0700 ~/.ssh && ssh-keyscan -p 2222 code.lazyworkhorse.net >> ~/.ssh/known_hosts 2>/dev/null
RUN --mount=type=ssh git clone --depth 1 --branch ${HONCHO_REF} ${HONCHO_REPO} /app
WORKDIR /app
ENV UV_COMPILE_BYTECODE=1
ENV UV_LINK_MODE=copy
ENV UV_PYTHON=/usr/local/bin/python3.13
RUN uv sync --frozen
# build stage — builds OpenConcho SPA
FROM node:22-bookworm AS openconcho-builder
ENV PNPM_HOME=/pnpm
ENV PATH=$PNPM_HOME:$PATH
RUN corepack enable && corepack prepare pnpm@latest --activate
WORKDIR /app
RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*
ARG OPENCONCHO_SHA=e490d911fcb27ee193558fd9a28856cde2057665
RUN git clone --depth 1 https://github.com/offendingcommit/openconcho.git /app && \
git -C /app fetch --depth 1 origin ${OPENCONCHO_SHA} && \
git -C /app checkout ${OPENCONCHO_SHA}
RUN pnpm install --frozen-lockfile
RUN pnpm --filter @openconcho/web build
# runtime stage — nginx + Honcho FastAPI
FROM python:3.13-slim-bookworm
# Install nginx and create runtime dirs before dropping permissions
RUN apt-get update && apt-get install -y --no-install-recommends nginx && \
rm -rf /var/log/nginx/* && \
rm -rf /var/lib/apt/lists/* && \
rm -f /etc/nginx/sites-enabled/default
# Patch nginx.conf: comment out "user www-data;" so nginx master stays as root
# (workers inherit root inside a container — fine for single-service isolation)
RUN sed -i 's/^user /# user /' /etc/nginx/nginx.conf
# Pre-create nginx runtime directories with proper ownership
RUN mkdir -p /var/lib/nginx/body /var/lib/nginx/proxy /var/lib/nginx/fastcgi \
/var/lib/nginx/uwsgi /var/lib/nginx/scgi /var/lib/nginx/proxy_temp \
/var/cache/nginx && \
chown -R root:root /var/lib/nginx /var/cache/nginx
# Honcho
COPY --from=honcho-builder /app /app
WORKDIR /app
ENV PATH="/app/.venv/bin:$PATH"
ENV HOME=/app
COPY config.toml /app/config.toml
# OpenConcho SPA
COPY --from=openconcho-builder /app/packages/web/dist /usr/share/nginx/html
# nginx config (proxies /v3/, /v2/ to Honcho on localhost:8000)
COPY honcho-nginx.conf /etc/nginx/conf.d/default.conf
EXPOSE 80
CMD ["bash", "-c", "nginx -g 'daemon off;' & fastapi run --host 127.0.0.1 --port 8000 src/main.py & python3 -m src.deriver & wait -n"]

View File

@@ -1,132 +0,0 @@
[app]
LOG_LEVEL = "INFO"
MAX_MESSAGE_SIZE = 25000
EMBED_MESSAGES = true
NAMESPACE = "honcho"
[db]
CONNECTION_URI = "postgresql+psycopg://honcho:honcho_pass@honcho-db:5432/honcho"
SCHEMA = "public"
POOL_SIZE = 10
MAX_OVERFLOW = 20
[auth]
USE_AUTH = false
[sentry]
ENABLED = false
[telemetry]
ENABLED = false
[webhook]
ENABLED = false
[cache]
ENABLED = true
URL = "redis://honcho-redis:6379/0"
[llm]
DEFAULT_MAX_TOKENS = 4096
# Embeddings via Ollama — bge-m3 provides 1024-dim
[embedding]
VECTOR_DIMENSIONS = 1024
MAX_INPUT_TOKENS = 8192
[embedding.model_config]
transport = "openai"
model = "bge-m3"
overrides = {base_url = "http://ollama:11434/v1", api_key = "ollama"}
# --- Deriver ---
[deriver]
ENABLED = true
WORKERS = 1
POLLING_SLEEP_INTERVAL_SECONDS = 5.0
FLUSH_ENABLED = true
[deriver.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
# --- Dialectic ---
[dialectic]
MAX_INPUT_TOKENS = 4096
SESSION_HISTORY_MAX_TOKENS = 8192
[dialectic.levels.minimal]
MAX_TOOL_ITERATIONS = 1
MAX_OUTPUT_TOKENS = 512
[dialectic.levels.minimal.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
[dialectic.levels.low]
MAX_TOOL_ITERATIONS = 3
[dialectic.levels.low.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
[dialectic.levels.medium]
MAX_TOOL_ITERATIONS = 2
[dialectic.levels.medium.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
[dialectic.levels.high]
MAX_TOOL_ITERATIONS = 4
[dialectic.levels.high.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
[dialectic.levels.max]
MAX_TOOL_ITERATIONS = 10
[dialectic.levels.max.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
# --- Summary ---
[summary]
ENABLED = true
MESSAGES_PER_SHORT_SUMMARY = 20
MESSAGES_PER_LONG_SUMMARY = 60
[summary.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
# --- Dream ---
[dream]
ENABLED = true
[dream.model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
[dream.deduction_model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
[dream.induction_model_config]
overrides = {base_url = "https://opencode.ai/zen/go/v1", api_key_env = "HONCHO_OPENAI_API_KEY"}
transport = "openai"
model = "deepseek-v4-flash"
# --- Peer Card ---
[peer_card]
ENABLED = true
# --- Vector Store ---
[vector_store]
TYPE = "pgvector"
# DIMENSIONS is deprecated — EMBEDDING.VECTOR_DIMENSIONS is authoritative

View File

@@ -1,52 +0,0 @@
server {
listen 80 default_server;
listen [::]:80 default_server;
server_name _;
root /usr/share/nginx/html;
index index.html;
# Honcho API proxy
location /v3/ {
proxy_pass http://127.0.0.1:8000;
proxy_http_version 1.1;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
location /v2/ {
proxy_pass http://127.0.0.1:8000;
proxy_http_version 1.1;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
# Honcho health
location /health {
proxy_pass http://127.0.0.1:8000;
proxy_http_version 1.1;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
# OpenAPI docs
location /openapi.json {
proxy_pass http://127.0.0.1:8000;
proxy_http_version 1.1;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
# SPA: fallback to index.html for client-side routing
location / {
try_files $uri $uri/ /index.html;
}
}

View File

@@ -1 +0,0 @@
CREATE EXTENSION IF NOT EXISTS vector;

View File

@@ -1,7 +0,0 @@
import shutil, os
src = "/opt/data/projects/gortium/compose/ai/compose_updated.txt"
dst = "/opt/data/projects/gortium/compose/ai/compose.yml"
print(f"Source exists: {os.path.exists(src)}, size: {os.path.getsize(src)}")
print(f"Dest exists: {os.path.exists(dst)}")
shutil.copy2(src, dst)
print(f"Copied, dest size: {os.path.getsize(dst)}")

View File

@@ -1,154 +0,0 @@
#!/bin/bash
# ── Hermes Workspace Combined Entrypoint ──
# Waits for the Hermes gateway container (hermes:8642) to become healthy,
# then starts the Hermes Workspace web UI in the foreground.
# Supports graceful shutdown via SIGTERM/SIGINT.
# ──────────────────────────────────────────
set -euo pipefail
# ── Configuration ──────────────────────────────────────────────
GATEWAY_HOST="${GATEWAY_HOST:-hermes}"
GATEWAY_PORT="${GATEWAY_PORT:-8642}"
GATEWAY_URL="http://${GATEWAY_HOST}:${GATEWAY_PORT}"
HEALTH_ENDPOINT="${HEALTH_ENDPOINT:-/health}"
MAX_RETRIES="${HEALTH_MAX_RETRIES:-60}"
RETRY_INTERVAL="${HEALTH_RETRY_INTERVAL:-2}"
WORKSPACE_DIR="${WORKSPACE_DIR:-/workspace}"
WORKSPACE_ENTRY="${WORKSPACE_ENTRY:-server-entry.js}"
PID_FILE="${PID_FILE:-/tmp/workspace.pid}"
# ── Logging ────────────────────────────────────────────────────
log_info() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] [INFO] $*"; }
log_warn() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] [WARN] $*"; }
log_error() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] [ERROR] $*"; }
# ── Graceful Shutdown ──────────────────────────────────────────
_workspace_pid=""
_shutting_down=false
cleanup() {
if [ "$_shutting_down" = true ]; then
return
fi
_shutting_down=true
log_info "Shutdown signal received, cleaning up..."
# Stop workspace process if running
if [ -n "$_workspace_pid" ] && kill -0 "$_workspace_pid" 2>/dev/null; then
log_info "Stopping workspace (PID: $_workspace_pid)..."
kill -TERM "$_workspace_pid" 2>/dev/null || true
# Give it time to shut down gracefully
local wait_sec=10
while kill -0 "$_workspace_pid" 2>/dev/null && [ "$wait_sec" -gt 0 ]; do
sleep 1
wait_sec=$((wait_sec - 1))
done
# Force kill if still running
if kill -0 "$_workspace_pid" 2>/dev/null; then
log_warn "Workspace did not shut down gracefully, force killing..."
kill -KILL "$_workspace_pid" 2>/dev/null || true
fi
fi
# Clean up PID file
[ -f "$PID_FILE" ] && rm -f "$PID_FILE"
log_info "Shutdown complete."
exit 0
}
# Trap termination signals for graceful shutdown
trap cleanup SIGTERM SIGINT
# ── Gateway Health Check ───────────────────────────────────────
wait_for_gateway() {
local url="${GATEWAY_URL}${HEALTH_ENDPOINT}"
local retries="$MAX_RETRIES"
local interval="$RETRY_INTERVAL"
local attempt=0
log_info "Waiting for Hermes gateway at ${GATEWAY_URL}..."
log_info "Max retries: ${retries}, interval: ${interval}s"
while [ "$attempt" -lt "$retries" ]; do
attempt=$((attempt + 1))
if curl -fsS "${url}" >/dev/null 2>&1; then
log_info "Gateway is healthy after ${attempt} attempt(s) (${GATEWAY_URL})"
return 0
fi
if [ "$attempt" -lt "$retries" ]; then
log_info "Gateway not ready yet (attempt ${attempt}/${retries}), retrying in ${interval}s..."
sleep "$interval"
fi
done
log_error "Gateway did not become healthy after ${retries} attempts (${retries * interval}s)"
return 1
}
# ── Workspace Startup ──────────────────────────────────────────
start_workspace() {
local entry="${WORKSPACE_DIR}/${WORKSPACE_ENTRY}"
if [ ! -d "$WORKSPACE_DIR" ]; then
log_error "Workspace directory not found: ${WORKSPACE_DIR}"
return 1
fi
if [ ! -f "$entry" ]; then
log_error "Workspace entry point not found: ${entry}"
return 1
fi
log_info "Starting Hermes Workspace web UI..."
log_info " Directory: ${WORKSPACE_DIR}"
log_info " Entry: ${entry}"
cd "$WORKSPACE_DIR"
# Start workspace in background so we can trap signals
exec node --max-old-space-size=2048 "${entry}" &
_workspace_pid=$!
echo "$_workspace_pid" > "$PID_FILE"
log_info "Workspace started (PID: ${_workspace_pid})"
# Wait for workspace process
wait "$_workspace_pid"
local exit_code=$?
log_info "Workspace exited with code ${exit_code}"
return "$exit_code"
}
# ── Main ───────────────────────────────────────────────────────
main() {
log_info "=== Hermes Workspace Combined Entrypoint ==="
log_info "Gateway: ${GATEWAY_URL}"
log_info "Workspace: ${WORKSPACE_DIR}/${WORKSPACE_ENTRY}"
log_info "PID file: ${PID_FILE}"
# Wait for gateway to be healthy
if ! wait_for_gateway; then
log_warn "Proceeding without confirmed gateway health..."
fi
# Start the workspace
start_workspace
local exit_code=$?
log_info "Entrypoint exiting with code ${exit_code}"
return "$exit_code"
}
# Run main; exit with its return code
main "$@"

View File

@@ -1,17 +0,0 @@
#!/usr/bin/env python3
"""Copy the updated txt file over the original yml file."""
import shutil
import os
src = "/opt/data/projects/gortium/compose/ai/compose_updated.txt"
dst = "/opt/data/projects/gortium/compose/ai/compose.yml"
# Check src exists
print(f"Source exists: {os.path.exists(src)}")
print(f"Source size: {os.path.getsize(src)} bytes")
print(f"Destination exists: {os.path.exists(dst)}")
# Copy
shutil.copy2(src, dst)
print(f"Copied {src} -> {dst}")
print(f"Destination size: {os.path.getsize(dst)} bytes")