KS
Killer-Skills

session-management — Categories.community

v1.0.0
GitHub

About this Skill

Ideal for Azure-focused AI Agents requiring structured session tracking and phase management in Python An Azure Web App in Python to crawl the content of SharePoint sites into OpenAI Vector Stores

karstenheld3 karstenheld3
[0]
[0]
Updated: 3/4/2026

Quality Score

Top 5%
33
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add karstenheld3/SharePoint-GPT-Middleware/session-management

Agent Capability Analysis

The session-management MCP Server by karstenheld3 is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion.

Ideal Agent Persona

Ideal for Azure-focused AI Agents requiring structured session tracking and phase management in Python

Core Value

Empowers agents to efficiently track EDIRD phases, manage session folders, and interact with OpenAI Vector Stores using Azure Web Apps, while maintaining comprehensive content analysis through NOTES.md, PROBLEMS.md, and PROGRESS.md files

Capabilities Granted for session-management MCP Server

Tracking EDIRD phases for project management
Automating session folder creation and organization
Integrating OpenAI Vector Stores for enhanced content analysis

! Prerequisites & Limits

  • Requires Azure Web App setup
  • Python environment necessary
  • Dependent on OpenAI Vector Stores and SharePoint site access
Project
SKILL.md
4.1 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

Session Management Guide

Phase Tracking

Sessions track EDIRD phases:

  • NOTES.md: "Current Phase" section with phase, last verb, gate status
  • PROGRESS.md: "Phase Plan" section with 5 phases and status

MUST-NOT-FORGET

  • Session folder location: [DEFAULT_SESSIONS_FOLDER]/_YYYY-MM-DD_[SessionTopicCamelCase]/
  • Default: [DEFAULT_SESSIONS_FOLDER] = [WORKSPACE_FOLDER] (override in !NOTES.md)
  • Required files: NOTES.md, PROBLEMS.md, PROGRESS.md
  • Lifecycle: Init → Work → Save → Resume → Finalize → Archive
  • Sync session PROBLEMS.md to project on /session-finalize
  • Phase tracking: NOTES.md has current phase, PROGRESS.md has full phase plan
  • STOP after session init: After creating session files, STOP and wait for user review. Do NOT implement session goal until explicitly requested. User must review and refine goals before work begins.

Session Lifecycle

  1. Init (/session-new): Create session folder with tracking files
  2. Work: Create specs, plans, implement, track progress
  3. Save (/session-save): Document findings, commit changes
  4. Resume (/session-load): Re-read session documents, continue work
  5. Finalize (/session-finalize): Sync findings to project files, prepare for archive

Session Folder Location

Base: [DEFAULT_SESSIONS_FOLDER] (default: [WORKSPACE_FOLDER], can be overridden in !NOTES.md)

Format: [DEFAULT_SESSIONS_FOLDER]/_YYYY-MM-DD_[SessionTopicCamelCase]/

Example: _PrivateSessions/_2026-01-12_FixAuthenticationBug/

Required Session Files

Use templates from this skill folder:

  • NOTES.md (NOTES_TEMPLATE.md): Key information, agent instructions, working patterns, large initial prompts (>120 tokens)
  • PROBLEMS.md (PROBLEMS_TEMPLATE.md): All problems to be addressed - initial prompts, questions, feature requests, bugs, strange behavior, investigation topics. Each problem gets a unique ID and tracks status (Open/Resolved/Deferred)
  • PROGRESS.md (PROGRESS_TEMPLATE.md): Task execution tracking - to-do list, done items, tried-but-not-used approaches

Key distinction:

  • NOTES.md = Context and reference information (static knowledge)
  • PROBLEMS.md = All topics requiring attention (dynamic problem list with IDs)
  • PROGRESS.md = Task execution status (what's being worked on)

Assumed Workflow

1. INIT: User initializes session (`/session-new`)
   └── Session folder, NOTES.md, PROBLEMS.md, PROGRESS.md created

2. PREPARE (one of):
   A) User prepares work manually
      └── Creates INFO / SPEC / IMPL documents, tracks progress
   B) User explains problem, agent assists
      └── Updates Problems, Progress, Notes → researches → creates documents

3. WORK: User or agent implements
   └── Makes decisions, creates tests, implements, verifies
   └── Progress and findings tracked continuously

4. SAVE: User saves session for later (`/session-save`)
   └── Everything updated and committed

5. RESUME: User resumes session (`/session-load`)
   └── Agent primes from session files, executes workflows in Notes
   └── Continue with steps 2-3

6. FINALIZE: User finalizes session (`/session-finalize`)
   └── Everything updated, committed, synced to project/workspace

7. ARCHIVE: User archives session
   └── Session folder moved to _Archive/

ID System

See [AGENT_FOLDER]/rules/devsystem-ids.md rule (always-on) for complete ID system.

Quick Reference:

  • Document: [TOPIC]-[DOC][NN] (IN, SP, IP, TP)
    • Example: CRWL-SP01, AUTH-IP01
  • Tracking: [TOPIC]-[TYPE]-[NNN] (BG = Bug, FT = Feature, PR = Problem, FX = Fix, TK = Task)
    • Example: SAP-BG-001, UI-PR-003, GLOB-TK-015
  • Topic Registry: Maintained in project NOTES.md

Session Init Template

NOTES.md

markdown
1# Session Notes 2 3## Session Info 4- **Started**: [DATE] 5- **Goal**: [Brief description] 6 7## Key Decisions 8 9## Important Findings 10 11## Workflows to Run on Resume

PROBLEMS.md

markdown
1# Session Problems 2 3## Open 4 5## Resolved 6 7## Deferred

PROGRESS.md

markdown
1# Session Progress 2 3## To Do 4 5## In Progress 6 7## Done 8 9## Tried But Not Used

Related Skills

Looking for an alternative to session-management or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication