memory-keeper

memory-keeper is an agent for coding agents from panaversity/agentfactory-business-plugins. It costs 61 tokens per session (1,047 once invoked), scanned A, original, Apache-2.0.

An agent that maintains work.local.md, a file used as the organisation’s shared memory. It records people, projects, decisions, terms, and other workplace context.

In plain words
What is it for?
It notices new people, projects, or terms; proposes additions; and supports memory updates, reviews, and checks for stale information.
Why use it?
It reduces the need to repeat workplace information and helps prevent important context from becoming outdated.

Agent

Part of the agentic-office plugin — 10 skills, 4 agents shipped together

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/panaversity/agentfactory-business-plugins/memory-keeper
Clone the repo
git clone --depth 1 https://github.com/panaversity/agentfactory-business-plugins

Or install agentic-office, the plugin that ships this one along with the rest of its 10 skills, 4 agents.

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 memory-keeper

README.md
[![agentmods](https://agentmods.dev/badge/agents/panaversity/agentfactory-business-plugins/memory-keeper.svg)](https://agentmods.dev/agents/panaversity/agentfactory-business-plugins/memory-keeper)
Your own site
<a href="https://agentmods.dev/agents/panaversity/agentfactory-business-plugins/memory-keeper"><img src="https://agentmods.dev/badge/agents/panaversity/agentfactory-business-plugins/memory-keeper.svg" alt="Measured on agentmods" height="20"></a>
Per session 61 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,047 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 $0.00061 $0.01047
Opus 5 $0.00030 $0.00524
Sonnet 5 $0.00012 $0.00209
Haiku 4.5 $0.00006 $0.00105

Measured 4d ago against content hash b10d8248508b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

memory-keeper 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.

agentic-office/agents/memory-keeper.md · 138 lines

How it starts

The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.

INVOCATION MODE

This agent has background: true which enables automatic invocation in Claude Code CLI when trigger conditions are detected.

In Cowork mode, this agent must be invoked via a Scheduled Task. See COWORK-SCHEDULED-TASKS.md for the recommended schedule and prompt. The agent's behavior and output are identical in both modes - only the invocation mechanism differs.

AGENT PURPOSE

Maintain the work.local.md file as the single source of organisational truth. After every significant interaction — a meeting, a decision, a new person encountered, a new project initiated, a new term used — the Memory Keeper proposes precise updates to work.local.md. Ensures workplace memory stays current rather than decaying into the outdated context that causes the Context Problem.

TRIGGER-BASED TASKS

TRIGGER: New person mentioned who is NOT in work.local.md

Immediately on detection: "I noticed you mentioned [Name] — they are not in workplace memory yet. Want me to add them? I can draft an entry based on what you have shared."

Draft entry using the workplace-context skill TYPE 1 format. Wait for confirmation before applying.

TRIGGER: New project name or codename used

"I noticed you referenced [Project name] — I do not have a record of this project. Would you like me to create a project entry in work.local.md?"

Draft entry using the workplace-context skill TYPE 2 format. Ask: priority, owner, current status, any known risks.

TRIGGER: New terminology used

"You used the term '[Term]' — I do not have a definition for this in the terminology dictionary. Want me to add it?"

Draft entry: term + definition + when to use + when not to use.

TRIGGER: Meeting completed

Within 2 hours of a meeting ending (coordinate with Meeting Intelligence Agent): Propose updates for:

  • Project status changes (if any project status shifted)
  • New decisions → add to decision log
  • New actions → add to action log / delegation log
  • New information about people (if anything was revealed about a person)
  • New terminology used in the meeting

Read the full file on GitHub · 138 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 · 138 lines · 61 tokens per session scan A b10d8248508b

Subscribe to this mod's changes

memory-keeper is an agent published in the GitHub repository panaversity/agentfactory-business-plugins (28 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 61 tokens to every session and 1,047 once invoked, about $0.0003 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

mirror-retro

Use this agent to run a retrospective on a completed feature, epic, or sprint. Mirror reads all artifacts in a feature folder (discovery, PRD, plan, architecture, build output, verification) and extracts what worked, what didn't, patterns to reuse, patterns to avoid, and proposes memory updates so the next cycle…

evolution-foundation/evo-nexus · 273 tokens

cpp-reviewer

Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.

affaan-m/ECC · 41 tokens

context

You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.

WrongStack/WrongStack · 0 tokens

portable-memory-parent-orchestrator

Top orchestrator for the portable-process-memory feature. Delegates to sync-transport (push/fetch folded into the verbs, plain-git, credential inheritance, offline-fail-safe) and event-fold (ownership events + the fail-closed divergence tripwire in the gate fold). Architect-only; coordinates portability/transport work…

seanrreid/RAD_framework · 77 tokens

context

Agent "context" from hannsxpeter/godpowers, covering scope, context, decisions, rules and workflows.

hannsxpeter/godpowers · 0 tokens

cpp-reviewer

Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.

DekaPrayoga/AurixAgent · 41 tokens