Troubleshooting
Plugin update not taking effect
After a plugin update from the Claude Code marketplace, commands from the old version may still be loaded. The fix is to clear the plugin cache and reinstall.
macOS / Linux:
# 1. Clear the stale cache
rm -rf ~/.claude/plugins/cache/stayinginsync-knowledge-graph
# 2. In Claude Code, run:
/plugin uninstall kmgraph
/reload-plugins
/plugin update stayinginsync
/plugin install kmgraph
/reload-plugins
Then fully quit and relaunch Claude Code, restart the MCP server (/mcp restart kmgraph), and verify with /kmgraph:init (select option 1 — Verify/upgrade).
Windows (PowerShell):
Remove-Item -Recurse -Force "$env:USERPROFILE\.claude\plugins\cache\stayinginsync-knowledge-graph"
Then follow the same /plugin uninstall → /plugin install steps above.
The marketplace may still show the older version number after /plugin update stayinginsync. That is expected — continue with the reinstall.
Commands do not appear in Claude Code autocomplete
- Verify the plugin is loaded: start Claude Code with
claude --plugin-dir /path/to/knowledge-graph - Commands use a colon, not a hyphen:
/kmgraph:init(correct),/knowledge-init(incorrect) - Restart Claude Code completely if commands still do not appear after the above
MCP server does not start
# Verify Node.js is installed (18+ required)
node --version
# Check the MCP server binary exists
ls mcp-server/dist/index.js
# Test the MCP server directly
./tests/test-mcp-direct.sh
If the binary is missing, rebuild it:
cd mcp-server && npm install && npm run build
Templates are not found
Verify that core/templates/ exists in the project directory and that templates were copied:
ls core/templates/
cp -r core/templates/. docs/templates/
Which lesson category should I use?
| Category | Use for |
|---|---|
architecture | System design decisions, component relationships |
process | Workflow improvements, tool configurations, procedures |
patterns | Reusable solutions discovered through experience |
debugging | Bug investigations, troubleshooting sessions, root cause analysis |
When uncertain: use debugging for problem-solving and process for workflow-related insights.
Is git required?
Git is recommended but not required. With git, the system automatically captures branch name, commit hash, and PR/issue numbers as lesson metadata. Without git, all features remain available — only automatic code linking is unavailable.
Still stuck?
Open an issue at github.com/technomensch/knowledge-graph/issues.