torbjorn/my-pal-mcp-server
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,151 Commits 1 Branch 0 Tags
main
Commit Graph

9 Commits

Author SHA1 Message Date
Fahad
3e27319e60 fix: reduced token usage, removed parameters from schema that CLIs never seem to use 2025-10-22 13:31:08 +04:00
Fahad
cc8a4dfd21 Overall savings should now be 50%+ tokens used 
perf: tweaks to schema descriptions, aiming to reduce token usage without performance degradation
 2025-10-01 22:39:12 +04:00
Fahad
4b202f5d1d feat: refactored and tweaked model descriptions / schema to use fewer tokens at launch (average reduction per field description: 60-80%) without sacrificing tool effectiveness 
Disabled secondary tools by default (for new installations), updated README.md with instructions on how to enable these in .env
run-server.sh now displays disabled / enabled tools (when DISABLED_TOOLS is set)
 2025-08-22 09:23:59 +04:00
Fahad
6921616db3 WIP: tool description / schema updates 2025-08-22 06:53:05 +04:00
Fahad
57200a8a2e Precommit updated to always perform external analysis (via _other_ model) unless specified not to. This prevents Claude from being overconfident and inadequately performing subpar precommit checks. 
Improved precommit continuations to be immediate
Workflow state restoration added between stateless calls
Fixed incorrect token limit check
 2025-08-20 15:19:01 +04:00
Fahad
adbc4af4a9 Update confidence enum values across workflow tools 
Added new confidence values (very_high, almost_certain) to all workflow tools
to provide more granular confidence tracking. Updated enum declarations in:
- analyze.py, codereview.py, debug.py, precommit.py, secaudit.py, testgen.py
- Updated debug.py's get_required_actions to handle new confidence values
- All tools now use consistent 7-value confidence scale
- refactor.py kept its unique scale (exploring/incomplete/partial/complete)

Also fixed model thinking configuration:
- Added very_high and almost_certain to MODEL_THINKING_PREFERENCES
- Set medium thinking for very_high, high thinking for almost_certain
- Updated prompts to clarify certain means 100% local confidence

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-06-28 00:04:35 +04:00
Fahad
bc447d4bcd Generic naming to work with Gemini CLI / Claude Code 2025-06-27 23:41:20 +04:00
Fahad
090931d7cf Fixed linebreaks 
Cleanup
Pass excluded fields to the schema builder directly
 2025-06-27 14:29:10 +04:00
Beehive Innovations
000d12dc3a Add secaudit tool for security auditing (#117) 
* WIP - working version

* Implement required methods
 2025-06-22 15:28:05 +04:00