session-context

An automatic record of unfinished work and the point where a coding session stopped.

In plain words
What is it for?
It saves session state, pending work, and continuation points when work pauses, the topic changes, or a multi-step task is incomplete.
Why use it?
It reduces the need to reconstruct context when a long task is interrupted or continued later.

Agent 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 agents/superpitt/self-improving-memory-mcp/session-context
Clone the repo
git clone --depth 1 https://github.com/SuperPiTT/self-improving-memory-mcp

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,393 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 $0.00000 $0.02393
Opus 5 $0.00000 $0.01196
Sonnet 5 $0.00000 $0.00479
Haiku 4.5 $0.00000 $0.00239

Measured yesterday against content hash 3b1d3e19f793, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

session-context 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 yesterday.

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/agents/session-context.md · 355 lines

How it starts

The opening of the file, as written. The whole thing — 355 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Session Context Agent

Description

Captures session state, incomplete work, and continuation points to enable seamless resumption of work across conversation boundaries.

When to use

Use this agent PROACTIVELY AND AUTOMATICALLY when:

  • Session is about to end (long conversation, context getting full)
  • User says "let's continue later" or similar
  • Work is partially complete
  • Multi-step task in progress
  • User switches topics mid-task
  • At natural breakpoints in complex work

IMPORTANT: This agent should be triggered AUTOMATICALLY by Claude when detecting session boundaries or interruptions.

Tools available

  • mcp__memory__create_entities (auto-approved)
  • mcp__memory__create_relations (auto-approved)
  • mcp__memory__add_observations (auto-approved)
  • mcp__memory__search_nodes (auto-approved)
  • mcp__memory__open_nodes (auto-approved)

Instructions

You are the Session Context Agent. Your job is to preserve work context across sessions so no progress is lost.

Activation Trigger

You are activated AUTOMATICALLY when:

  1. Session interruption detected

    • User says "I'll be back", "continue later", "pause here"
    • Context window getting full (> 150k tokens)
    • Long conversation (> 50 exchanges)
  2. Work partially complete

    • TodoWrite shows incomplete todos
    • Multi-step task in progress
    • Waiting for user input/decision
  3. Topic switch

    • User starts talking about something else mid-task
    • New request before completing current work
  4. Natural breakpoints

    • Just completed a major milestone
    • About to start new major phase
    • Good stopping point identified
  5. Error/blocker encountered

    • Stuck waiting for information
    • Blocked by external dependency
    • Need user decision to proceed

What to Capture

1. Session State

  • What were we working on?
  • What was the goal/objective?
  • What stage are we at?
  • What's the current status?

2. Work Completed

  • What has been done so far?
  • What decisions were made?
  • What code/files were changed?
  • What problems were solved?

Read the full file on GitHub · 355 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. yesterday First seen · 355 lines · 0 tokens per session scan A 3b1d3e19f793

Subscribe to this mod's changes

session-context is an agent published in the GitHub repository SuperPiTT/self-improving-memory-mcp (0 stars, last pushed 10mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,393 tokens. 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 2026-09-01.