Description
Agentmemory-hooks explains how the agentmemory plugin captures memory automatically during a coding session. For Claude Code, the plugin registers lifecycle hooks that observe tool use, prompts and session boundaries, and write observations without you calling a save command for routine work. Session start and end frame each unit of work, which lets a handoff resume it later. Tool-use hooks record what changed and why, which is the raw material for later recall and recap. A prompt-submit hook captures intent, and a pre-compact hook preserves context before the host trims it. A post-commit hook links commits to sessions, which powers the commit-context and commit-history skills. The skill also explains that Hermes and OpenClaw use their own integrations, and that installing shared skills or connecting MCP alone does not turn on automatic capture. To check that capture works, run a real turn and confirm it creates an observation, which you can watch in the viewer at port 3113 by default. The plugin is installed through the Claude Code plugin marketplace.…
Highlights
- Automatic capture across the session lifecycle
- Tool-use, prompt-submit and pre-compact hooks
- Post-commit hook links commits to sessions
- Viewer shows observations at the default port
Ratings & Reviews
No ratings yet
Be the first to rate agentmemory-hooks.
People also visit
Claude Code Infrastructure Showcase
by diet103 - A remarkably innovative approach to working
configure-ecc
Run the conversational ECC setup wizard inside the current harness: inventory
continuous-learning
[DEPRECATED - use continuous-learning-v2] Legacy v1 stop-hook skill extractor
agentmemory-hooks FAQ
How does memory get captured?
The Claude Code plugin's lifecycle hooks record observations from tool use, prompts and session boundaries.
Do I need to save memories myself?
Not for routine work. The hooks write observations automatically, though you can still call memory tools directly.
Does connecting MCP turn on capture?
No. The skill says that installing shared skills or connecting MCP alone does not enable automatic capture.
How do I check that it works?
Run a real turn and confirm that an observation is created, then view it in the viewer at localhost port 3113.
How do I install it?
For Claude Code, add the rohitg00/agentmemory marketplace and install the agentmemory plugin from it. Run npx skills add https://github.com/rohitg00/agentmemory --skill agentmemory-hooks for the skill.
Is it free?
The listing shows the skill as free.
What security audits has it passed?
skills.sh lists three audits: Gen Agent Trust Hub Pass, Socket Pass and Snyk Warn. A Warn or Fail is worth reading before you install.
What are the best agentmemory-hooks alternatives?
Popular alternatives to agentmemory-hooks include Claude Code Infrastructure Showcase, git-guardrails-claude-code and configure-ecc. See all agentmemory-hooks alternatives