← Back to Dashboard

🎯 Priority Work Queue

What CultureSherpa needs to work on, in order

Updated: 2025-12-06 02:34
Agent Success Rate: 35%

📋 Executive Summary

Agent can only fix dependencies automatically. Security/linting/testing/docs require LLM integration or human intervention.

Blocked: security, linting, testing, docs, todo Working: dependencies

📊 Prioritized Issues

#1

Security

CRITICAL

Hardcoded credentials and API keys need environment variable replacement

Found: 285 Fixed: 0
⚠️ Blocker: Agent heuristics cannot safely modify credential patterns
💡 To Fix: Create .env template, use python-dotenv, agent can then auto-migrate
❌ BLOCKED
#2

Linting

MEDIUM

Ruff warnings: unused imports, f-strings, unused variables

Found: 30 Fixed: 0
⚠️ Blocker: Agent heuristics not implemented for Ruff auto-fix
💡 To Fix: Add 'ruff --fix' as pre-commit hook or agent fix strategy
❌ BLOCKED
#3

Testing

MEDIUM

Missing tests for key modules

Found: 15 Fixed: 0
⚠️ Blocker: Test generation requires LLM or template system
💡 To Fix: Use pytest fixtures + LLM to generate test cases
❌ BLOCKED
#4

Docs

LOW

Functions missing docstrings

Found: 30 Fixed: 0
⚠️ Blocker: Docstring generation requires understanding function purpose
💡 To Fix: Extract function signatures, use LLM to generate docstrings
❌ BLOCKED
#5

Todo

LOW

TODO placeholder docstrings and comments

Found: 4532 Fixed: 0
⚠️ Blocker: Each TODO requires contextual understanding
💡 To Fix: Already fixed 1,566 in a previous commit - continue pattern
❌ BLOCKED
#6

Type_Hints

LOW

Missing type annotations

Found: 30 Fixed: 0
⚠️ Blocker: Type inference requires code analysis
💡 To Fix: monkeytype or pytype for automatic type inference
❌ BLOCKED

🚀 Recommended Next Actions

P0
Review security issues
285 hardcoded credentials need attention before any deployment
SECURITY_AUDIT.md
P1
Enable ruff auto-fix
Quick win - will resolve 30 linting issues
ruff check --fix .
P1
Create .env template
Foundation for migrating hardcoded credentials
.env.example

🧠 Agent Learning Insights

Patterns Discovered: 5
Heuristic Success Rate: 20% success rate overall
Most Useful Patterns:
  • File opened without context manager (1645 instances)
  • Broad exception catch (508 instances)
  • Blocking sleep in async code (153 instances)
📄 Raw JSON Manifest 📝 Markdown Report