Context Management
Managing context effectively for better results
Effective context management ensures Verdent has the right information at the right time while avoiding performance degradation from context overload.
What You'll Learn
- Understanding context windows and their limits
- Selecting files strategically for optimal context
- Recognizing and responding to context overload
- When to reset context for better performance
Understanding Context Windows
Verdent's context window size depends on the model being used. Extended context (1M) is handled automatically when needed.
Available Models:
| Provider | Model | Context |
|---|---|---|
| Anthropic | claude-opus-4.5 | 200K |
| Anthropic | claude-sonnet-4.5 | 200K |
| Anthropic | claude-haiku-4.5 | 200K |
| gemini-3-pro-preview | 1M | |
| gemini-3-flash-preview | 1M | |
| OpenAI | gpt-5 | 272K |
| OpenAI | gpt-5-codex (Beta) | 272K |
| Zhipu AI | GLM 4.7 | 200K |
| Moonshot | kimi-k2-turbo | 262K |
What's Included in Context:
- All messages in conversation
- File contents loaded into context
- Tool outputs and responses
- System prompts and instructions
Viewing Context Usage
Monitor your current context window usage through the Context Remaining indicator in the middle panel. This shows how much context space remains for the current conversation.
Context Commands:
| Command | Description |
|---|---|
/compact | Compress conversation history to free up context space |
/clear | Clear conversation and start fresh |
Use /compact when context is running low but you want to continue the same task. Use /clear or start a new task when switching to unrelated work.
Strategic File Selection
Be strategic about file selection to optimize context usage.
Start with fewer files and add more only when needed. Verdent can always read additional files during the conversation.
Use @-mentions for explicit file inclusion:
@filename.jsVerdent auto-loads related files, but @-mentions ensure precise control. Only include files directly relevant to the current task.
- Include only files that need to be modified or referenced
- Reference existing patterns instead of loading example files
- For large codebases, work on one module at a time
- Let Verdent read files as needed rather than preloading
Recognizing Context Overload
Signs:
- Less accurate or incomplete responses
- Missing important details from earlier in conversation
- Confused about recent changes or context
Action: Start a fresh task when quality degrades
Signs:
- Noticeably slower response times
- Longer processing delays
Action: Consider starting a new task
Signs:
- Requests to clarify information already provided
- Forgetting patterns established earlier
- Redundant questions about project structure
Action: Reset context by starting a new task
If Verdent's responses become vague or repetitive, context overload may be occurring. Start a new task to restore full performance.
When to Reset Context
- Noticeably slower response times
- Less accurate or inconsistent responses
- Verdent forgetting earlier context
Action: Start fresh task when quality degrades
- Switching between unrelated features
- After completing one task and moving to next
- After memory-intensive tasks (large refactorings)
Action: New task for new major work
- After committing completed features
- Between logical checkpoints
- Following test-verify-commit cycles
Action: Commit → test → new task
Best Practice: Complete atomic unit of work → test → commit → start fresh task.