save-session

A command that summarizes the current coding conversation and saves the summary to a file, so it can be used in a later session.

In plain words
What is it for?
Use it before ending a work session or switching conversations. You can also include the next task for the following session.
Why use it?
It reduces the context lost when starting a fresh conversation and records what was done, what changed, and what remains.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/ggprompts/tfe/save-session
Clone the repo
git clone --depth 1 https://github.com/GGPrompts/TFE

Made for: Claude Code.

Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,147 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00011 $0.01147
Opus 5 $0.00005 $0.00574
Sonnet 5 $0.00002 $0.00229
Haiku 4.5 $0.00001 $0.00115

Measured 3d ago against content hash acafc055c7d1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

save-session scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/save-session.md · 175 lines

How it starts

The opening of the file, as written. The whole thing — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Save Session: Summarize and Export Current Work Session

You are creating a concise summary of the current work session and saving it to a file, enabling the user to continue in a completely fresh conversation or resume later with minimal context loss.

Note: This is different from the built-in /compact command which compresses the conversation in-place. This command EXPORTS the summary to a file for external use with reload scripts.

Command Arguments

The user may provide an optional argument after /save-session:

  • If provided: Use it as the opening prompt for the next session
  • If empty: Just create summary without a specific next task

Examples:

  • /save-session - Just create summary
  • /save-session Let's implement syntax highlighting in the preview - Summary + specific next goal
  • /save-session Debug the tree view expansion issue - Summary + next debugging task

Your Task

Create a comprehensive but concise summary of our conversation that captures:

1. Session Overview

  • What we accomplished
  • What features were implemented
  • What problems were solved
  • Current state of the project

2. Technical Details

Files Modified:

  • List all files created or modified
  • Include key changes made to each file
  • Note any new functions, types, or significant refactors

Architecture Decisions:

  • Any important architectural choices
  • New patterns introduced
  • Refactoring decisions

Current State:

  • Where we left off
  • What's working
  • What's in progress
  • Any known issues or TODOs

3. Context for Continuation

Next Steps:

  • What should be worked on next
  • Any blockers or dependencies
  • Suggested priorities

Important Context:

  • Key insights or discoveries
  • Things to remember
  • Gotchas or caveats

Summary Format

Use this structure for the summary:

# Session Summary - [Date/Time]

## What We Accomplished
- [Bullet point list of main achievements]

## Files Modified
### Created:
- `path/to/file.ext` - Description of what it does

### Modified:
- `path/to/file.ext` - What changed and why

## Key Technical Details
- [Important implementation details]
- [Architecture decisions]
- [New patterns or approaches]

## Current State
- ✅ Working: [What's functional]
- 🔄 In Progress: [What's partially done]
- ⚠️ Issues: [Known problems]

## Next Steps
1. [Priority 1]
2. [Priority 2]
3. [Priority 3]

## Important Context
- [Key insights to remember]
- [Gotchas or warnings]
- [Dependencies or requirements]

---

## NEXT SESSION GOAL

[If user provided argument to /save-session, include it here with clear formatting]

**User wants to work on:**
[Their specified goal/task]

[If no argument provided, use "No specific goal set - ready to continue general development"]

Read the full file on GitHub · 175 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 175 lines · 11 tokens per session scan A acafc055c7d1

Subscribe to this mod's changes

save-session is a command published in the GitHub repository GGPrompts/TFE (20 stars, last pushed 2mo ago), licensed MIT. It adds 11 tokens to every session and 1,147 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.