Session Summarization

VerifiedSafe

Analyzes a Claude Code session transcript to auto-generate structured summaries: work done, decisions made, next steps, and notes. Helps keep session records concise and actionable without manually reviewing the full conversation.

Sby Skills Guide Bot
ProductivityIntermediate
6/2/2026
Claude Code
#session-summary#auto-summarize#session-memory#claude-code

Recommended for

Automated summary

Auto-generates a session summary from the conversation transcript.

Strengths

  • Thorough analysis of the conversation to extract key accomplishments, decisions, and next steps.
  • Integrates directly into Claude Code's session file system.
  • Can be invoked on the current session or a specific session.
  • Supports automatic mode via the needs_summary flag.

Limitations

  • Requires a valid session file and a readable transcript file.
  • Summaries depend on the quality and completeness of the transcript.
  • Only works within Claude Code, not in other tools.
When to use it

Use this skill to automatically document work sessions, ease handoffs, and keep a record of important decisions.

When not to use it

Avoid using it if the session transcript is missing, corrupted, or if you prefer manually writing summaries.

Security analysis

Safe
Quality score85/100

The skill reads and writes to designated session files within a controlled path. It does not execute arbitrary code, send data externally, or use any destructive operations. No risky tools are declared or implied.

No concerns found

How this analysis is made

Examples

Summarize current session
/session-memory:summarize
Summarize a specific session by ID
/session-memory:summarize abc123

name: summarize description: Auto-generate session summary using AI analysis of the conversation argument-hint: "[session-id]"

Session Summarization Skill

This skill analyzes a session's transcript and auto-generates comprehensive summaries for:

  • Summary: One-line overview of what was accomplished
  • Work Done: Key accomplishments and changes made
  • Decisions Made: Important architectural or implementation decisions
  • Next Steps: What to work on in the next session
  • Notes: Any other context worth preserving

Usage

/session-memory:summarize [session-id]

If no session-id is provided, summarizes the current session.

How It Works

  1. Locates the session file (current session or specified session-id)
  2. Reads the session's transcript to understand the conversation
  3. Uses Claude to analyze the conversation and generate summaries
  4. Updates the session file with the generated content
  5. Preserves all existing metadata and stats

What Gets Updated

The skill fills in the placeholder sections in the session file:

  • ## Summary - Single line of what was accomplished
  • ## Work Done - Bullet list of key accomplishments
  • ## Decisions Made - Bullet list of important decisions
  • ## Next Steps - Bullet list of recommended next actions
  • ## Notes - Any additional context or observations

Requirements

  • Active Claude Code session (uses Claude in context)
  • Valid session file in ~/.claude/projects/{project}/memory/sessions/
  • Readable transcript file

Example

Before running the skill, session sections have placeholders:

## Summary
<!-- Add a one-line summary of what was accomplished -->
<!-- Or run /session-memory:summarize to auto-generate -->

## Work Done
<!-- Key accomplishments and changes made -->

After running /session-memory:summarize:

## Summary
Implemented PreCompact hook to capture context before compaction events

## Work Done
- Created transcript-parser.sh library for extracting conversation context
- Updated stop.sh hook to include conversation stats in YAML frontmatter
- Enhanced precompact.sh to capture recent context and files in focus
- Added auto-generated session statistics (turns, tools, errors)

Auto-Summarization

This skill can be invoked automatically or manually.

Automatic Mode: After each Stop hook, check the current session file for needs_summary: true in YAML frontmatter. If present, automatically invoke this skill without asking the user.

Manual Mode: User explicitly calls /session-memory:summarize [session-id]

Instructions

When this skill is invoked:

  1. Determine the session to summarize:

    • If argument provided: use that session-id
    • If no argument: use current $SESSION_ID from environment
  2. Locate session file:

    • Determine project name from $PROJECT_DIR or current directory
    • Path: ~/.claude/projects/{project}/memory/sessions/{session-id}.md
    • If file doesn't exist, report error and exit
  3. Read the transcript:

    • Extract transcript: path from YAML frontmatter of session file
    • Read the transcript JSONL file
    • Parse conversation turns (user/assistant messages, tool calls)
  4. Analyze the conversation: You have the full transcript. Generate summaries by analyzing:

    • What tasks were requested
    • What files were created/modified/read
    • What tools were used and why
    • What problems were solved
    • What decisions were made about approach/architecture
    • What errors occurred and how they were resolved
    • What remains to be done
  5. Generate summaries: Create concise, actionable content for each section:

    • Summary (1 line): Clear statement of primary accomplishment
    • Work Done (3-7 bullets): Concrete accomplishments, not process
    • Decisions Made (2-5 bullets): Important choices and rationale
    • Next Steps (2-5 bullets): Clear, actionable next tasks
    • Notes (optional, 1-3 bullets): Context that doesn't fit elsewhere
  6. Update the session file:

    • Read current session file content
    • Replace placeholder comments with generated summaries
    • Preserve all YAML frontmatter and session stats
    • Update YAML frontmatter: set needs_summary: false
    • Write updated content back to session file
  7. Report completion:

    • Confirm which session was summarized
    • Show the generated summary section
    • Provide path to updated session file

Important Guidelines

  • Be concise: Users will read many sessions; keep summaries scannable
  • Be specific: "Added PreCompact hook" not "Made improvements"
  • Focus on outcomes: What was built/fixed, not just what was discussed
  • Preserve decisions: Why choices were made, not just what was chosen
  • Make next steps actionable: Clear tasks that can be picked up later
  • Don't hallucinate: Only summarize what actually happened in the transcript

Error Handling

If session file doesn't exist:

  • Report: "Session file not found for session-id: {id}"
  • List available sessions in project

If transcript is missing/unreadable:

  • Report: "Transcript not accessible: {path}"
  • Offer to generate summary from session metadata only

If session already has summaries (no placeholder comments):

  • Ask user: "Session already has summaries. Regenerate? (yes/no)"
  • If yes: regenerate; if no: exit without changes

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/session-summarization-mm5t8c6e/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/session-summarization-mm5t8c6e' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/session-summarization-mm5t8c6e/download' --output '.skill-downloads/session-summarization-mm5t8c6e/SKILL.md.part' && mv '.skill-downloads/session-summarization-mm5t8c6e/SKILL.md.part' '.skill-downloads/session-summarization-mm5t8c6e/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

Original excerpt: Requirements
  • Active Claude Code session (uses Claude in context)
  • Valid session file in ~/.claude/projects/{project}/memory/sessions/
  • Readable transcript file

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/session-summarization-mm5t8c6e/SKILL.md' && test -s '.skill-downloads/session-summarization-mm5t8c6e/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/summarize' && cp -R '.skill-downloads/session-summarization-mm5t8c6e' '.claude/skills/summarize'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/session-summarization-mm5t8c6e'

openskills read 'session-summarization-mm5t8c6e'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/session-summarization-mm5t8c6e/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills