agentsclimarketplace

Session memory

Skill RhythmMittal19/claude-config/skills/session-memory

Ultimate Claude Code configuration — 9 agents, 19 skills, 13 templates. Researched from 850+ repos.

Install
npx -y skills add RhythmMittal19/claude-config --skill session-memory

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Bridges Code tab sessions with the Memory MCP server. Use at session start to load relevant project memories, and at session end to store new discoveries. Activates when working on any project to maintain continuity across sessions.

SKILL.md

1.9 KB, 387 tokens by cl100k_base, as published. Nobody here has run it

Session Memory Bridge

Purpose

The Memory MCP server persists facts across Claude Desktop conversations. This skill helps Claude Code sessions (Code tab) leverage that same memory.

At Session Start

  1. Check if the memory MCP server has entries for this project:
    echo "Project: $(basename $(pwd))"
    
  2. Look for learnings.md in the project root
  3. Look for .claude/session-handoff.md from previous sessions
  4. Apply all relevant context before starting work

During Work — What to Remember

Store these types of facts (for the Memory MCP in Desktop chat later):

  • Architecture decisions: "Project X uses JWT auth with refresh tokens in httpOnly cookies"
  • Conventions discovered: "In project X, all API calls go through services/ layer"
  • Gotchas found: "In project X, the test DB must be seeded before running integration tests"
  • Dependencies: "Project X requires Node 20+ and MongoDB 7+"
  • Current state: "Project X has auth module complete, payment module in progress"

At Session End

  1. Generate a summary of what was done
  2. Update learnings.md with new discoveries
  3. If ending a long session, create .claude/session-handoff.md
  4. Suggest facts to store in Memory MCP (user can do this in Desktop chat)

Session Handoff Format

## Session Handoff — [date]
### Done: [list of completed work]
### Decisions: [architecture/design choices made]
### State: [what's working, what's in progress]
### Issues: [bugs found, blockers, tech debt]
### Next: [prioritized next steps]
### Files: [key files modified]
### Memory Suggestions: [facts to tell Memory MCP in Desktop chat]

Gives 0 of the 12 instructions most memory context skills give in 387 tokens

Counted across 674 of the 847 authors here whose files we hold, read 2026-08-06

  • inform the user when setup is completein 21 of 674, across 6 files
  • confirm the draft with the user before writingin 21 of 674, across 6 files
  • update the agent skills block in place if it existsin 21 of 674, across 6 files
  • present findings to the userin 20 of 674, across 5 files
  • write the three docs files from seed templatesin 20 of 674, across 5 files
  • ask the user about each decision one at a timein 19 of 674, across 4 files
  • edit CLAUDE.md if it existsin 18 of 674, across 3 files
  • explore current repo statein 18 of 674, across 3 files
  • do not overwrite user edits to surrounding sectionsin 18 of 674, across 3 files
  • back up the original file before overwritingin 16 of 674, across 8 files
  • keep the memory index under 200 linesin 15 of 674
  • Provide actionable steps and verificationin 13 of 674, across 2 files

Said here and by no other author read

  • look for learnings.md in the project root
  • look for previous session-handoff files
  • apply all relevant context before starting work
  • store discovered architecture decisions and conventions
  • store discovered gotchas, dependencies, and current state
  • update learnings.md with new discoveries at session end

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.