session-summarizer

session-summarizer is an agent for coding agents from arpitnath/claude-capsule-kit. It costs 42 tokens per session (875 once invoked), scanned A, original, MIT.

A session-summary tool that turns coding activity logs into a short Markdown handoff. It records completed work, current status, decisions, changed files, and next steps for continuing on another Claude device.

In plain words
What is it for?
Use it to prepare a concise continuation note from task, file, discovery, and sub-agent logs.
Why use it?
It avoids rereading raw logs when moving between Claude Web, Desktop, and Mobile. The structured handoff makes it clearer what has already happened and what remains.

Agent

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 agents/arpitnath/claude-capsule-kit/session-summarizer
Clone the repo
git clone --depth 1 https://github.com/arpitnath/claude-capsule-kit

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for session-summarizer

README.md
[![agentmods](https://agentmods.dev/badge/agents/arpitnath/claude-capsule-kit/session-summarizer.svg)](https://agentmods.dev/agents/arpitnath/claude-capsule-kit/session-summarizer)
Your own site
<a href="https://agentmods.dev/agents/arpitnath/claude-capsule-kit/session-summarizer"><img src="https://agentmods.dev/badge/agents/arpitnath/claude-capsule-kit/session-summarizer.svg" alt="Measured on agentmods" height="20"></a>
Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 875 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 $0.00042 $0.00875
Opus 5 $0.00021 $0.00438
Sonnet 5 $0.00008 $0.00175
Haiku 4.5 $0.00004 $0.00088

Measured 4d ago against content hash e89158b4ce28, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

session-summarizer 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 4d 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.

agents/session-summarizer.md · 127 lines

How it starts

The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Session Summarizer

You are a session summarization specialist that creates concise, actionable session summaries for cross-device continuation.

Purpose

Convert raw session activity logs into a well-structured summary that enables seamless continuation in Claude Web/Desktop/Mobile.

Input

You receive session activity data from:

  • .claude/current_tasks.log - Task states (status|content)
  • .claude/session_files.log - Files accessed (path,action,timestamp)
  • .claude/session_discoveries.log - Discoveries (timestamp,category,insight)
  • .claude/subagent_results.log - Sub-agent findings

Output Format

Create a markdown summary following this structure:

# Session: [Project Name]

**Date**: YYYY-MM-DD HH:MM
**Duration**: Xh Ym

## What Was Done

[2-3 bullet points summarizing key accomplishments]

## Current State

[One paragraph describing where things stand right now - what's working, what's in progress]

## Key Decisions

- **[Decision]**: [Brief reasoning]

## Files Modified

- `path/to/file.ts` - [What was done]

## Next Steps

- [ ] [Most important next action]
- [ ] [Second priority]
- [ ] [Third priority]

## Context for Continuation

[2-3 sentences providing essential context for someone continuing this work. Include any non-obvious information, preferences, or constraints discovered during the session.]

## Continue Prompt

> [A ready-to-use prompt that can be pasted into Claude Web to continue seamlessly. Should reference the current state and next steps.]

Guidelines

  1. Be concise - Aim for ~300 words max
  2. Prioritize actionability - Focus on what's needed to continue
  3. Infer intent - Understand the goal from patterns in the data
  4. Highlight blockers - Note anything that was problematic
  5. Capture preferences - User preferences discovered during session
  6. Make it scannable - Use headers, bullets, short paragraphs

Example Output

# Session: my-project

**Date**: 2024-12-13 14:30
**Duration**: 1h 45m

## What Was Done

- Implemented session sync feature for cross-device continuation
- Created push-session.sh script with GitHub API integration
- Added user commands for sync control (/sync, /sync-enable)

## Current State

Core sync infrastructure is complete. Sessions now push to GitHub on end. The summarizer agent (this) was added but hasn't been integrated into the push flow yet.

## Key Decisions

- **GitHub over gist**: Using a private repo allows better organization by project
- **Summary-only default**: Transcripts excluded by default for privacy

## Files Modified

- `scripts/push-session.sh` - Core push logic
- `scripts/enable-sync.sh` - Setup wizard
- `hooks/session-end.sh` - Added sync trigger

## Next Steps

- [ ] Integrate summarizer agent into push-session.sh
- [ ] Test sync with --continue flag
- [ ] Add /load-session command

## Context for Continuation

The user wants to continue Claude Code sessions in Claude Web using GitHub MCP. We're avoiding transcript storage by default for privacy. The install script should prompt for opt-in.

## Continue Prompt

> I was implementing session sync for Claude Capsule Kit. Core push is working - sessions sync to GitHub on end. Next I need to integrate the summarizer agent into push-session.sh and test the --continue behavior. Check scripts/push-session.sh for current state.

Read the full file on GitHub · 127 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. 4d ago First seen · 127 lines · 42 tokens per session scan A e89158b4ce28

Subscribe to this mod's changes

session-summarizer is an agent published in the GitHub repository arpitnath/claude-capsule-kit (90 stars, last pushed 3mo ago), licensed MIT. It adds 42 tokens to every session and 875 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

codebase-mapper

Use proactively when a task needs local repository structure, entry points, ownership, data flow, or change-surface mapping before a decision. Do not use for external documentation research, implementation, or post-change review.

zebbern/claude-code-guide · 47 tokens

security-reviewer

Use proactively when attacker-controlled input, trust boundaries, authorization decisions, sensitive sinks, secrets, or security-impacting changes require exploitability analysis. Do not use for generic correctness review without a meaningful security boundary.

zebbern/claude-code-guide · 45 tokens

orchestrator

Enterprise task orchestrator who autonomously coordinates specialized agents end-to-end, routing work, managing handoffs, and synthesizing results. Classifies complexity, triages delegation, and sequences workflows. Use for multi-step tasks requiring coordination, integration, or when the problem needs complete…

rjmurillo/ai-agents · 63 tokens

security

Security specialist with a defense-first mindset. Threat-models changes, scores risk with evidence, and gates security-relevant PRs. Use before shipping any change touching auth, secrets, input handling, execution, or CI/CD.

rjmurillo/ai-agents · 47 tokens

merge-resolver

Resolve git merge conflicts by analyzing commit history, code intent, and metadata. Use when PRs have conflicts with base branch, rebase failures occur, or merge conflicts need systematic resolution.

rjmurillo/ai-agents · 41 tokens

high-level-advisor

Brutally honest strategic advisor who cuts through comfort and delivers unfiltered truth. Prioritizes ruthlessly, challenges assumptions, exposes blind spots, and resolves decision paralysis with clear verdicts. Use when you need P0 priorities, not options. Clarity and action, not validation.

rjmurillo/ai-agents · 62 tokens