A procedure for maintaining durable information in a shared memory file. It is used only when someone explicitly asks the agent to remember, change, or forget something for later.

Use it to read existing memory, make a minimal update, or remove a remembered preference or fact when explicitly requested.

It keeps long-term notes separate from one-off conversations and limits updates to concise, factual information without secrets.

memory — Skill for Claude Code and Codex by SawyerHood/middleman

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.

Any agent

npx skills add SawyerHood/middleman --skill memory

Clone the repo

git clone --depth 1 https://github.com/SawyerHood/middleman

Made for: Claude Code, Codex.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/sawyerhood/middleman/memory/github.svg)](https://agentmods.dev/skills/sawyerhood/middleman/memory)

Your own site

<a href="https://agentmods.dev/skills/sawyerhood/middleman/memory"><img src="https://agentmods.dev/badge/skills/sawyerhood/middleman/memory/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/sawyerhood/middleman/memory"><img src="https://agentmods.dev/badge/skills/sawyerhood/middleman/memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 262 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00028	$0.00262
Opus 5.5	$0.00011	$0.00105
Sonnet 5.5	$0.00006	$0.00052
Haiku 4.5	$0.00003	$0.00026

Measured 2mo ago against content hash 8a1ae8b739e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

memory 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 2mo 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.

apps/backend/src/swarm/skills/builtins/memory/SKILL.md · 33 lines

What it actually says

Persistent Memory Workflow

Use this skill when the user explicitly asks to:

remember something for later,
update previously remembered facts/preferences, or
forget/remove stored memory entries.

Do not write memory for normal one-off requests.

File location

Persistent memory files are stored at ${SWARM_DATA_DIR}/memory/<agentId>.md.
In this runtime, use ${SWARM_MEMORY_FILE} (also shown in your loaded context).

Steps

Read the current memory file with read before changing it.
Apply minimal edits:
- prefer edit for targeted changes,
- use write only for full rewrites.
Keep entries concise, factual, and durable.
Never store secrets (passwords, API keys, tokens, private keys) or highly sensitive personal data.
If the request is ambiguous, ask a clarifying question before writing.
After updating memory:
- manager: confirm the update to the user via speak_to_user,
- worker: report the update back to the manager via send_message_to_agent.

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.

2mo ago First seen · 33 lines · 28 tokens per session scan A 8a1ae8b739e8

Subscribe to this mod's changes

memory is a skill published in the GitHub repository SawyerHood/middleman (185 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 28 tokens to every session and 262 once invoked, about $0.0001 per session on Opus 5.5. 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.

Claude Code Memory and context agents ai-agents

memory next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
memorySawyerHood/middleman	Skill	185	5mo ago	Apache-2.0	A	28	Claude Code, Codex
hindsight-localvectorize-io/hindsight	Skill	59k	yesterday	MIT	A	32	Claude Code, Codex
agentmemory-hooksrohitg00/agentmemory	Skill	30k	yesterday	Apache-2.0	A	42	Claude Code
agentmemory-agentsrohitg00/agentmemory	Skill	30k	yesterday	Apache-2.0	A	46	Claude Code
handoffrohitg00/agentmemory	Skill	30k	yesterday	Apache-2.0	A	55	Claude Code
memory-contextvectorize-io/hindsight	Skill	59k	yesterday	MIT	A	47	Cursor
memory-retainvectorize-io/hindsight	Skill	59k	yesterday	MIT	A	46	Cursor