memory-pipeline

What This Skill Does

The memory-pipeline is a comprehensive agent memory and performance framework designed to evolve OpenClaw agents from stateless responders into persistent, knowledge-aware assistants. It bifurcates into two distinct operational layers: the Memory Pipeline (for off-session knowledge management) and the Performance Routine (for in-session execution guardrails).

By leveraging structured fact extraction, knowledge graph construction, and multi-stage lifecycle hooks, the skill ensures your agent retains critical context, identifies conflicting data, and enforces strict operational parameters. It reduces context bloat through automated result compression and ensures long-term consistency via mandatory after-action reviews.

Installation

To install this skill, run the following command in your terminal: clawhub install openclaw/skills/skills/bodii88/memory-pipeline-0-1-0

After installation, ensure you have at least one valid LLM API key (OpenAI, Anthropic, or Gemini) configured in your environment. The scripts will auto-detect your workspace and provide the necessary intelligence to process session data into a persistent knowledge base.

Use Cases

• Long-term Project Management: Maintain state across multiple coding sessions, ensuring the agent remembers architectural decisions and pending tasks.
• Agent Consistency: Use the briefing injection to enforce specific personas, tone, or operational constraints every time the agent starts.
• Knowledge Consolidation: Automatically synthesize facts from fragmented daily logs and disorganized session transcripts into a coherent knowledge graph.
• Execution Quality Assurance: Prevent unsafe tool usage or excessive token consumption through automated output compression and strict tool-call filtering.

Example Prompts

1. "Run the full memory pipeline now to consolidate all facts from yesterday's development session."
2. "Generate a briefing for my next coding session that highlights active projects and critical unresolved technical constraints."
3. "Check the current knowledge graph for any contradictions regarding our current API authentication strategy."

Tips & Limitations

• Dependency Management: Ensure your environment has sufficient write permissions to the workspace directory for the script to generate logs and BRIEFING.md files.
• API Usage: Since this skill performs frequent LLM calls for extraction and summarization, be mindful of your API provider's rate limits and costs.
• Maintenance: Periodically review the knowledge graph structure to ensure extracted facts are relevant; the quality of your output is directly tied to the clarity of your input transcripts.