ai-space/opencode-working-memory
3
0
Fork 0
mirror of https://github.com/sdwolf4103/opencode-working-memory.git synced 2026-07-17 12:56:42 +02:00
Code Issues Packages Projects Releases Wiki Activity

docs: add token-usage guidance for session patterns

Browse Source
This commit is contained in:
Ralph Chang
2026-02-18 16:57:40 +08:00
parent 2edb643667
commit 7b33b51c1e
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+7
View File
@@ -176,6 +176,13 @@ The plugin exposes these tools to your OpenCode agent:
- ✅ Pressure-aware compression
- ✅ Zero configuration, works immediately
## Does working-memory system increase token usage? It depends.
- **If you use the DCP plugin and regularly start new sessions:** this plugin can add some overhead, because each new session starts fresh and the memory system setup may not pay off before the session ends.
- **If you stay in one long-running session and often hit token limits/auto-compaction:** this plugin is a strong fit, because repeated compaction can make the agent forget goals, active folders, and where critical files are.
The plugin solves this by preserving high-value memory (goal/progress/context + working memory signals) so your agent can inherit context from its "previous self" with minimal token cost, reducing repeated mistakes and the extra tokens those mistakes consume.
## Configuration (Optional)
The plugin works great with zero configuration. To customize behavior, modify the constants at the top of `index.ts`. See the [Configuration Guide](docs/configuration.md) for all tunable options.