agentleFS
maxwell2732maxwell2732/claudecode-research-harness-workflow/skills/session/SKILL.md

session

Skill42 starsChanged 6 days agoMITView on GitHub

Unified session management window. Handles initialization, memory, state all-in-one. Explicit /session invocation only — sub-skills handle auto-delegation. Use when managing Claude Code sessions, /session command. Do NOT load for: app user sessions, login state, authentication features.

18 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use debug, release and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AnthropicSkill

    debug

    Structured debugging session — reproduce, isolate, diagnose, and fix. Trigger with an error message or stack trace, "this works in staging but not prod", "something broke after the deploy", or when behavior diverges from expected and the cause isn't obvious.

    Also in 3 projects that use this

  • MicrosoftSkill

    release

    Use ONLY when the user explicitly asks to cut/publish a Huabu desktop release or tag a version. Performs an outward action — bumps the version in apps/desktop/package.json, commits + annotated-tags vX.Y.Z, and pushes the branch + tag. That tag push auto-builds the macOS/Windows installers and publishes a GitHub Release. Provides `status`, `bump`, `tag-push`, and a one-command `run` (bump → tag-push).

    Also in 4 projects that use this

  • GoogleSkill

    tdd

    Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

    Also in 3 projects that use this

The file

Files
SKILL.md

Session Skill (Unified)

Consolidates all session-related functionality into one skill.

Usage

/session              # Show available options
/session list         # Show active sessions
/session inbox        # Check incoming messages
/session broadcast "message"  # Send message to all sessions

Subcommands

/session list - List Active Sessions

Shows all active Claude Code sessions in the current project.

📋 Active Sessions

| Session ID | Status | Last Activity |
|------------|--------|---------------|
| abc123     | active | 2 min ago     |
| def456     | idle   | 15 min ago    |

/session inbox - Check Inbox

Checks for incoming messages from other sessions.

📬 Session Inbox

| From | Time | Message |
|------|------|---------|
| abc123 | 5m ago | "Ready for review" |
| def456 | 10m ago | "API implementation done" |

/session broadcast "message" - Broadcast Message

Sends a message to all active sessions.

/session broadcast "Review complete, ready for merge"

Capabilities

FeatureDescriptionReference
InitializationStart new session, load contextSee ../session-init/SKILL.md
MemoryPersist learnings across sessionsSee ../session-memory/SKILL.md
State ControlResume/fork session based on flagsSee references/session-control.md
CommunicationCross-session messagingSee ../session-state/SKILL.md

Memory Optimization (CC 2.1.49+)

From Claude Code 2.1.49 onward, memory usage when resuming sessions has been reduced by 68%.

Best Practices for Long Session Management

WorkloadRecommended Strategy
Normal implementationResume with --resume every 1–2 hours
Large-scale refactoringSplit by feature unit → use --resume per session
Parallel tasksRun in parallel with /work all; use --resume for long sessions
On memory warningImmediately resume with --resume (faster than before)

Automatic Session Naming (CC 2.1.41+)

Running /rename without arguments automatically generates a session name from the conversation context. Makes session identification easier in workflows that involve long sessions or frequent --resume.

Efficient Workflow Example

# Implementation phase 1
claude "Implement authentication"
# → 1 hour later

# Resume session (memory-efficient)
claude --resume "Add password reset feature"
# → 1 hour later

# Resume again
claude --resume "Add tests"

Memory Management Recommendations

RecommendationReason
Actively resume sessions68% memory reduction makes resuming cheap
Resume periodicallyOrganizes context and maintains focus
Split by feature unitBreak large tasks into smaller resumable chunks
Leverage Plans.mdSmooth handoff on resume

Memory efficiency has greatly improved — actively leverage session resumption.

Codex 0.123.0 session shell / terminal Fix

In Codex 0.123.0, stale proxy env is less likely to be restored from the shell snapshot, and Unicode / dead-key input in VS Code WSL terminal and keyboard input have been fixed in the main binary. The harness does not add a proxy snapshot scrubber or key input wrapper — it inherits the main binary fixes automatically.


When to Use

  • Session initialization (/harness-init)
  • Session resume/fork (/work --resume, /work --fork)
  • Memory persistence (automatic)
  • Cross-session communication (/session broadcast)

Execution Flow

1. Session Initialization

/harness-init
    ↓
├── Load project context
├── Initialize session.json
├── Load previous session memory (if exists)
└── Display session status

2. Session Control (from /work)

/work --resume
    ↓
├── Check session.json exists
├── Load session state
└── Continue from last checkpoint

/work --fork
    ↓
├── Create new session branch
├── Copy relevant context
└── Start fresh with context

3. Memory Persistence

Session end
    ↓
├── Extract learnings (gotchas, patterns)
├── Update .claude/memory/*.md
└── Prepare handoff summary

4. Cross-Session Communication

/session broadcast "message"
    ↓
├── Find active sessions
├── Write to session.events.jsonl
└── Notify all sessions

Files Managed

FilePurpose
.claude/state/session.jsonCurrent session state
.claude/state/session.events.jsonlEvent log for cross-session communication
.claude/memory/*.mdPersistent memory files

Migration Note

This skill consolidates:

  • session-init → Session initialization
  • session-memory → Memory persistence
  • session-control → Resume/fork control
  • session-state → State management & communication

The individual skills are deprecated but still work for backward compatibility.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.