First Version

honcho/.env.example

@@ -0,0 +1,72 @@
+# Honcho — environment configuration
+# Copy this file to .env and fill in the values.
+#
+#   cp .env.example .env
+#
+# Lines marked [REQUIRED] must be set before starting.
+# Lines marked [OPTIONAL] have sensible defaults and can be left empty.
+
+# ── LLM Provider ─────────────────────────────────────────────────────────────
+#
+# Honcho uses an LLM internally for:
+#   - extracting conclusions from conversation turns  (deriver)
+#   - building peer representations / user cards       (deriver)
+#   - session summarisation                            (deriver)
+#   - dialectic/peer_chat queries                      (api)
+#
+# The deriver will fail silently (no memory extraction) if no key is provided.
+# The API itself will still respond; only background memory-building stops.
+
+# Option A — OpenAI (default)
+# All model slots default to gpt-4o-mini / text-embedding-3-small.
+LLM_OPENAI_API_KEY=sk-...                    # [REQUIRED for default setup]
+
+# Option B — OpenRouter (OpenAI-compatible, access to many models)
+# Uncomment and set DERIVER_MODEL_CONFIG__OVERRIDES__BASE_URL in the section below.
+# LLM_OPENAI_API_KEY=sk-or-...
+
+# Option C — Ollama (fully local, no data leaves the machine)
+# Requires Ollama running on the host and a model that supports function calling.
+# Uncomment the OLLAMA block below.
+
+# ── Model overrides (optional) ────────────────────────────────────────────────
+#
+# Leave empty to use the defaults (OpenAI gpt-4o-mini / text-embedding-3-small).
+# Uncomment and edit to point at a different provider or model.
+
+# OpenRouter example:
+# DERIVER_MODEL_CONFIG__TRANSPORT=openai
+# DERIVER_MODEL_CONFIG__MODEL=openai/gpt-4o-mini
+# DERIVER_MODEL_CONFIG__OVERRIDES__BASE_URL=https://openrouter.ai/api/v1
+
+# Ollama example (fully local):
+# LLM_OPENAI_API_KEY=ollama
+# DERIVER_MODEL_CONFIG__TRANSPORT=openai
+# DERIVER_MODEL_CONFIG__MODEL=llama3.3:70b
+# DERIVER_MODEL_CONFIG__OVERRIDES__BASE_URL=http://host.docker.internal:11434/v1
+# LLM_EMBEDDING_API_KEY=ollama
+# LLM_EMBEDDING_BASE_URL=http://host.docker.internal:11434/v1
+# LLM_EMBEDDING_MODEL=nomic-embed-text
+
+# ── Database ──────────────────────────────────────────────────────────────────
+# Credentials for the built-in PostgreSQL container.
+# Change these if you connect an external DB.
+
+POSTGRES_USER=honcho
+POSTGRES_PASSWORD=honcho          # [RECOMMENDED] change in production
+POSTGRES_DB=honcho
+
+# ── Redis ─────────────────────────────────────────────────────────────────────
+# No configuration needed for the built-in Redis container.
+# To use an external Redis, override CACHE_URL in docker-compose.yml.
+
+# ── API exposure ──────────────────────────────────────────────────────────────
+# Port the Honcho API listens on. Default: 8000.
+HONCHO_PORT=8000
+
+# API auth token. Leave empty for unauthenticated local access.
+# If set, pass it as Bearer token from personal-agent's plugin config (api_key field).
+# HONCHO_AUTH_TOKEN=
+
+# ── Sentry / observability (optional) ────────────────────────────────────────
+# SENTRY_DSN=