Spaces:
Sleeping
Sleeping
File size: 2,831 Bytes
116524e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 | # Dockerfile.ace — OpenClaw + ACE (self-improving agent)
#
# Extends the OpenClaw image with Python 3.12 and the ACE framework
# pre-installed, so the agent can learn from past sessions automatically.
#
# Build (two-step, from the openclaw repo root):
#
# # 1. Build the base OpenClaw image
# docker build -t openclaw:base .
#
# # 2. Extend with ACE
# docker build -t openclaw:local -f Dockerfile.ace .
#
# Then set OPENCLAW_IMAGE=openclaw:local in your .env file.
#
# IMPORTANT: You must pass your LLM API key through docker-compose so
# ace-learn can call the reflection model. Add to docker-compose.yml
# under the gateway service's environment section:
#
# ACE_MODEL: "bedrock/us.anthropic.claude-sonnet-4-20250514-v1:0"
# # Plus one of: AWS_BEARER_TOKEN_BEDROCK, ANTHROPIC_API_KEY,
# # OPENROUTER_API_KEY, or LITELLM_API_KEY
ARG OPENCLAW_IMAGE=openclaw:base
FROM ${OPENCLAW_IMAGE}
USER root
# ---------------------------------------------------------------------------
# Python 3.12 + uv
# ---------------------------------------------------------------------------
# uv — fast Python package & version manager
RUN curl -LsSf https://astral.sh/uv/install.sh | env UV_INSTALL_DIR=/usr/local/bin sh
# Install Python 3.12 to a shared location
# (bookworm ships 3.11; uv downloads an official standalone build)
ENV UV_PYTHON_INSTALL_DIR=/opt/python
RUN uv python install 3.12
# ---------------------------------------------------------------------------
# ACE framework
# ---------------------------------------------------------------------------
RUN rm -rf /opt/ace \
&& git clone --depth 1 https://github.com/Kayba-ai/agentic-context-engine.git /opt/ace \
&& cd /opt/ace \
&& uv sync --no-dev --extra claude-code --python 3.12 \
&& uv pip install boto3 --python .venv/bin/python \
&& chown -R node:node /opt/ace
# Wrapper script: writes output to the persistent workspace volume
# so the skillbook survives container restarts.
RUN printf '#!/bin/bash\n\
set -euo pipefail\n\
export OPENCLAW_HOME="${OPENCLAW_HOME:-$HOME/.openclaw}"\n\
OUTPUT_DIR="$OPENCLAW_HOME/workspace/skills/kayba-ace"\n\
mkdir -p "$OUTPUT_DIR"\n\
cd /opt/ace\n\
exec .venv/bin/python examples/openclaw/kayba-ace/learn_from_traces.py \\\n\
--output "$OUTPUT_DIR" \\\n\
"$@"\n' > /usr/local/bin/ace-learn \
&& chmod +x /usr/local/bin/ace-learn
# Note: Do NOT set OPENCLAW_HOME here — the base image derives it from
# $HOME (~/.openclaw). Setting it explicitly causes double-nesting.
# ---------------------------------------------------------------------------
# Restore original user and working directory
# ---------------------------------------------------------------------------
USER node
WORKDIR /app
|