Stats
Actions
Tags
From agent-context-graph-claude
Checks, explains, and troubleshoots Agent Context Graph capture from Claude Code hooks into Memgraph-backed graph connectors using CLI doctor and hook smoke tests.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agent-context-graph-claude:agent-context-graphThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this skill when the user asks whether Claude Code activity is being captured, how Agent Context Graph hooks work, or why graph facts are missing.
Use this skill when the user asks whether Claude Code activity is being captured, how Agent Context Graph hooks work, or why graph facts are missing.
Claude Code Plugin -> Claude Code Runtime Adapter -> Event Protocol -> Graph Connector -> Memgraph.
The plugin only installs runtime hook wiring. It must not assign graph meaning. Graph connectors decide which normalized events matter.
Run the single CLI doctor first. It checks the same Python environment that the hook command uses:
agent-context-graph doctor --runtime claude-code --connector skills-graph
If doctor is not available, use the strict hook smoke:
printf '{"hook_event_name":"Stop","session_id":"doctor"}' \
| agent-context-graph hook run claude-code --connector skills-graph --strict
Do not check skills_graph with system python3; agent-context-graph may be installed in an isolated uv tool or pipx environment.
If skill usage is missing, inspect whether the session actually read or invoked a skill. Search/list results can be surfacing, not proven usage.
npx claudepluginhub memgraph/ai-toolkit --plugin context-graphGives agents low-token codebase context by querying a graphify knowledge graph instead of reading raw files. Use for research, impact analysis, and agent-loop integration.
Explains Claude Code internals: hooks, skills, headless mode, and debugging. Use when hooks don't fire, skills don't trigger, or you need to design an extension.
Audits agent lifecycles in Claude Code plugins, validating execution capabilities against configs via 8 semantic audits on alignments, skill loading, contracts, prompts, tools, dead agents, and patterns. Writes reports to .claude/audits/.