customable/claude-mem
1
0
Fork 0
Code Issues 1 Pull requests 5 Projects Releases 163 Packages 2 Wiki Activity Actions

feat: batch processing and performance improvements #90

Merged
jonas.hanisch merged 2 commits from feat/batch-processing-and-performance into main 2026-01-22 21:37:14 +00:00
Conversation 0 Commits 2 Files changed 9 +409 -173
jack commented 2026-01-22 21:05:22 +00:00
Owner
Copy link

Summary

  • Add batch observation processing to reduce API call overhead (configurable via CLAUDE_MEM_BATCH_SIZE, default: 5)
  • Add conversation history limiting to prevent token explosion (keep first 2 + last 10 messages max)
  • Add 120s request timeout for Mistral API calls to prevent hanging
  • Fix settings validation bugs (boolean handling, provider normalization)

Changes

Batch Processing

  • New buildBatchObservationPrompt() in src/sdk/prompts.ts for processing multiple tool events per API call
  • MistralAgent collects observations and processes them in batches instead of one-by-one
  • Configurable batch size via CLAUDE_MEM_BATCH_SIZE setting

Performance Fixes

  • Both MistralAgent and SDKAgent now limit conversation history to prevent exponential token growth
  • Added 120s timeout for Mistral API requests

Settings Validation Fixes

  • Accept actual boolean values, not just string "true"/"false"
  • Add 'openai' to valid providers list
  • Normalize 'anthropic' → 'claude' for backwards compatibility

Test plan

  • Verify batch processing works with MistralAgent (check logs for "Processing batch of X observations")
  • Verify conversation history doesn't grow beyond 12 messages
  • Test settings API accepts boolean values
  • Test settings API accepts 'openai' and 'anthropic' providers

🤖 Generated with Claude Code

## Summary - Add batch observation processing to reduce API call overhead (configurable via `CLAUDE_MEM_BATCH_SIZE`, default: 5) - Add conversation history limiting to prevent token explosion (keep first 2 + last 10 messages max) - Add 120s request timeout for Mistral API calls to prevent hanging - Fix settings validation bugs (boolean handling, provider normalization) ## Changes ### Batch Processing - New `buildBatchObservationPrompt()` in `src/sdk/prompts.ts` for processing multiple tool events per API call - MistralAgent collects observations and processes them in batches instead of one-by-one - Configurable batch size via `CLAUDE_MEM_BATCH_SIZE` setting ### Performance Fixes - Both MistralAgent and SDKAgent now limit conversation history to prevent exponential token growth - Added 120s timeout for Mistral API requests ### Settings Validation Fixes - Accept actual boolean values, not just string "true"/"false" - Add 'openai' to valid providers list - Normalize 'anthropic' → 'claude' for backwards compatibility ## Test plan - [ ] Verify batch processing works with MistralAgent (check logs for "Processing batch of X observations") - [ ] Verify conversation history doesn't grow beyond 12 messages - [ ] Test settings API accepts boolean values - [ ] Test settings API accepts 'openai' and 'anthropic' providers 🤖 Generated with [Claude Code](https://claude.com/claude-code)
feat: batch processing and performance improvements
All checks were successful
CI / build (pull_request) Successful in 21s
Details
CI / validate-plugin (pull_request) Successful in 2s
Details
4385a45184 
- Add batch observation processing (CLAUDE_MEM_BATCH_SIZE setting, default: 5)
- Add buildBatchObservationPrompt() for processing multiple tool events per API call
- Add conversation history limiting (keep first 2 + last 10 messages) to prevent token explosion
- Add 120s request timeout for Mistral API calls
- Fix settings validation to accept boolean values (not just strings)
- Add 'openai' to valid providers and normalize 'anthropic' → 'claude'

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
feat: auto-register hooks in settings.json during install
All checks were successful
CI / build (pull_request) Successful in 21s
Details
CI / validate-plugin (pull_request) Successful in 2s
Details
c554ab7d8d 
- Add registerHooks() function to smart-install.js
- Reads hooks.json from plugin and merges into user's settings.json
- Supports CLAUDE_CONFIG_DIR environment variable
- Creates backup of settings.json if parsing fails
- Non-blocking: hook registration failure doesn't stop plugin installation

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
jonas.hanisch deleted branch feat/batch-processing-and-performance 2026-01-22 21:37:14 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels   
good first issue

Good for newcomers

  
has-pr

Issue has an associated pull request

  
help wanted

Community contributions welcome

  
idea

Idea or concept - not yet ready for implementation

  
priority
critical
Urgent issue requiring immediate attention

  
priority
high
High priority issue

  
priority
low
Low priority issue

  
priority
medium
Medium priority issue

  
status
blocked
Blocked by external dependency

  
status
in-progress
Currently being worked on

  
status
needs-review
Awaiting code review

  
status
ready
Ready to be worked on

  
type
bug
Something isn't working

  
type
docs
Documentation improvements

  
type
enhancement
Improvement to existing functionality

  
type
feature
New feature or request

  
type
refactor
Code refactoring without functional changes

No labels
good first issue
has-pr
help wanted
idea
priority
critical
priority
high
priority
low
priority
medium
status
blocked
status
in-progress
status
needs-review
status
ready
type
bug
type
docs
type
enhancement
type
feature
type
refactor
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
customable/claude-mem!90
No description provided.