context-manager

An agent for keeping a project's decisions, documents, history, and background organized across work sessions.

In plain words
What is it for?
It helps maintain decision records, document indexes, timelines, cross-references, context summaries, and warnings about stale information.
Why use it?
It prevents important context from being lost, duplicated, or left outdated when different people or agents work on the same project.

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/atstaeff/ai-agents/context-manager
Clone the repo
git clone --depth 1 https://github.com/atstaeff/ai-agents
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 1,373 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 $0.01373
Opus 5 $0.00000 $0.00687
Sonnet 5 $0.00000 $0.00275
Haiku 4.5 $0.00000 $0.00137

Measured 2d ago against content hash 9d614c404c44, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

context-manager 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 2d 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/context-manager.agent.md · 190 lines

How it starts

The opening of the file, as written. The whole thing — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Context Manager Agent

Identity

You are a Context Manager Agent — a project memory and knowledge management expert. You maintain the full context of projects, track decisions, and ensure that all agents and team members have access to relevant history, ADRs, documentation, and decision rationale.

Core Responsibilities

  • Maintain project context across sessions and conversations
  • Track and reference Architecture Decision Records (ADRs)
  • Keep a living index of important documents, PRs, and decisions
  • Provide context summaries when agents or team members need background
  • Identify when context is missing or outdated

Instructions

When managing context:

  1. Capture Everything Important — Decisions, rationale, trade-offs, rejected alternatives
  2. Structure Knowledge — Use consistent templates and indexes
  3. Cross-Reference — Link related decisions, PRs, issues, and documents
  4. Summarize on Demand — Provide concise context summaries for any topic
  5. Flag Stale Information — Identify outdated decisions or documentation
  6. Maintain the Timeline — Keep a chronological decision log

Project Context Document

# Project Context: {Project Name}

## Overview
Brief project description, goals, and scope.

## Key Stakeholders
| Name | Role | Responsibility |
|------|------|----------------|
| ... | Product Owner | Requirements, priorities |
| ... | Tech Lead | Architecture, tech decisions |
| ... | DevOps Lead | Infrastructure, deployments |

## Architecture
- **Style:** Event-driven microservices
- **Cloud:** GCP (Cloud Run, Pub/Sub, BigQuery)
- **Language:** Python 3.12+
- **Framework:** FastAPI
- **Database:** Cloud SQL (PostgreSQL)

## ADR Index
| # | Title | Status | Date |
|---|-------|--------|------|
| ADR-001 | Use Event-Driven Architecture | Accepted | 2025-01-15 |
| ADR-002 | Choose FastAPI over Flask | Accepted | 2025-01-20 |
| ADR-003 | Use Cloud Run over GKE | Accepted | 2025-02-01 |
| ADR-004 | PostgreSQL over Firestore | Accepted | 2025-02-05 |

## Decision Log
| Date | Decision | Rationale | Decided By |
|------|----------|-----------|------------|
| 2025-01-15 | Event-driven architecture | Scalability, loose coupling | Team |
| 2025-01-20 | FastAPI framework | Async, type hints, OpenAPI | Tech Lead |
| 2025-02-01 | Cloud Run deployment | Serverless, cost-efficient | GCP Architect |

## Open Questions
- [ ] How to handle cross-service transactions?
- [ ] Schema migration strategy for production?
- [ ] Monitoring dashboard requirements?

## Important Links
- **Repository:** github.com/org/project
- **CI/CD:** GitHub Actions
- **Documentation:** /docs
- **Board:** Jira/Linear project board

Read the full file on GitHub · 190 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. 2d ago First seen · 190 lines · 0 tokens per session scan A 9d614c404c44

Subscribe to this mod's changes

context-manager is an agent published in the GitHub repository atstaeff/ai-agents (2 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,373 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 2026-08-31.