Comparison
Apothem vs raw OpenCode config
Hand-maintaining OpenCode config, commands, skills, and agents versus rendering them from Apothem's shared profile.
OpenCode reads JSON config plus native commands, skills, and agents discovery
directories. Apothem renders the JSON config with an instructions file list
and converts shared cohorts into native directories where OpenCode supports
them.
Effort comparison
| Task | Raw hand-config | Apothem-managed |
|---|---|---|
| Author rules | Hand-edit config and support prompts | Rendered or referenced from the shared profile |
| Author commands, skills, agents | Maintain each OpenCode directory by hand | Converted from shared Apothem cohorts |
| Keep cohorts aligned with other harnesses | Manual, per tool | Automatic on apothem update |
| Install / remove the surface | Manual | apothem install / apothem uninstall |
| Verify the config is correct | Manual inspection | apothem verify --harness opencode |
What Apothem manages
The opencode adapter writes opencode.json, converts Apothem slash commands to
OpenCode Markdown command files, copies skills into the native skills surface,
and adapts helper definitions as Markdown subagents. Rules, templates, and hook
prompts remain under Apothem's support subtree; rules are referenced from the
native instructions list.
Migration path
- Run
apothem profile initto create the shared profile if you have not yet. - Move existing OpenCode rules, commands, skills, and agents into the shared corpus.
- Run
apothem install --harness opencodeto render the native config and discovery directories. - Run
apothem verify --harness opencodeto confirm the config landed. - Use
apothem update --harness opencodefor every later change.