Agent Squad Python Guide
2FastLabs/agent-squad
Map of the agent-squad Python framework for async multi-agent orchestration: which agent, classifier, storage and tool provider to pick, and the pitfalls to avoid.
Store, recall, and forget durable facts with CAO memory — user preferences, project conventions, decisions, and corrections that should persist across sessions and agents.
Install Cao Memory
$ npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install awslabs/cli-agent-orchestrator cao-memory --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/awslabs/cli-agent-orchestrator.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/cao-memory .claude/skills/cao-memory && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "cao-memory" agent skill from https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memory into .claude/skills/cao-memory/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cao-memory", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memoryType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install awslabs/cli-agent-orchestrator cao-memory --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/awslabs/cli-agent-orchestrator.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/cao-memory .agents/skills/cao-memory && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "cao-memory" agent skill from https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memory into .agents/skills/cao-memory/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cao-memory", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install awslabs/cli-agent-orchestrator cao-memory --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/awslabs/cli-agent-orchestrator.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/cao-memory .cursor/skills/cao-memory && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "cao-memory" agent skill from https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memory into .cursor/skills/cao-memory/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cao-memory", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/awslabs/cli-agent-orchestrator.git --path skills/cao-memory--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install awslabs/cli-agent-orchestrator cao-memory --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/awslabs/cli-agent-orchestrator.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/cao-memory .gemini/skills/cao-memory && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "cao-memory" agent skill from https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memory into .gemini/skills/cao-memory/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cao-memory", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install awslabs/cli-agent-orchestrator cao-memoryInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/awslabs/cli-agent-orchestrator.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/cao-memory .github/skills/cao-memory && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "cao-memory" agent skill from https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memory into .github/skills/cao-memory/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cao-memory", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install awslabs/cli-agent-orchestrator cao-memory --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/awslabs/cli-agent-orchestrator.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/cao-memory .opencode/skills/cao-memory && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "cao-memory" agent skill from https://github.com/awslabs/cli-agent-orchestrator/tree/main/skills/cao-memory into .opencode/skills/cao-memory/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cao-memory", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
cao-memory- GitHub stars
- 1.4k
- Token cost
- ~1.4k tokens
- SKILL.md length
- 650 words
- Files
- 1
- Skills in repo
- 14
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Store, recall, and forget durable facts with CAO memory — user preferences, project conventions, decisions, and corrections that should persist across sessions and agents.
- Works in 4 steps: Recall before asking. The answer may… → Store the instant you learn something… → One fact per memory, 1–2 sentences.… → …
- Tasks that involve Multi-agent orchestration
- SKILL.md covers Scopes and types, Recall — check memory BEFORE…, Store — save anything worth… and Forget — remove what's wrong…, plus 2 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Cao Memory is an agent skill from awslabs/cli-agent-orchestrator, published by the product's own GitHub organization. Store, recall, and forget durable facts with CAO memory — user preferences, project conventions, decisions, and corrections that should persist across sessions and agents. Use proactively to check memory before asking the user, and to save anything worth remembering. Distinct from any provider-native memory.
Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Agent Workflows, covering Multi-agent orchestration. It works with Amazon DynamoDB and Model Context Protocol. The repository describes itself as: Multi-agent orchestration for AI coding CLIs — Claude Code, Kiro, Codex, and more, coordinated in isolated tmux sessions. The licence is Apache-2.0.
When your agent uses it
- Tasks that involve Multi-agent orchestration
Example prompts
- “/cao-memory”
Workflow steps
4 steps, taken from the first numbered list in SKILL.md.
- Recall before asking. The answer may already be stored.
- Store the instant you learn something durable — corrections, conventions,
- One fact per memory, 1–2 sentences. Conclusions, not conversation.
- Pick the right scope: user-wide → global; this repo → project.
What it can do on your machine
Read from SKILL.md and the folder at commit b29f40a. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Cao Memory loads about 1.4k tokens when it runs. Until then it costs about 80 tokens; SKILL.md has 650 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from awslabs/cli-agent-orchestrator at commit b29f40a, republished under its Apache-2.0 licence (© awslabs). 650 words, ~1,366 tokens.
Download SKILL.mdSave it as
.claude/skills/cao-memory/SKILL.md (or your agent's skills folder).- name
- cao-memory
- description
- Store, recall, and forget durable facts with CAO memory — user preferences, project conventions, decisions, and corrections that should persist across sessions and agents. Use proactively to check memory before asking the user, and to save anything worth remembering. Distinct from any provider-native memory.
CAO Memory
CAO gives every agent a shared, persistent memory. A fact you store in one session is available to a brand-new agent in a later session — even on a different provider. Use it so the user never has to repeat themselves.
These are CAO's cross-provider memory tools (memory_store, memory_recall,
memory_forget), exposed by the CAO MCP server. They are distinct from any
provider-native memory the CLI tool may have.
Scopes and types
Every memory has a scope (where it applies) and a type (what kind of fact it is).
| Scope | Applies to | Use for |
|---|---|---|
project (default) | This repo / working directory | Conventions, architecture, build rules |
global | Every project | User identity, durable cross-project preferences |
federated | Every project on this machine | Reusable, repo-independent lessons worth sharing across all your work (rejects credentials) |
session | This run only | Short-lived task context |
agent | This agent role | Role-specific working notes |
Types: project (default), user (who the user is / preferences), feedback
(corrections and how-to-work guidance), reference (pointers to docs, tickets, URLs).
Recall — check memory BEFORE asking the user
At the start of a task, and whenever you're about to ask the user something they may have already told you, search memory first.
memory_recall(query="database widgets endpoint testing")Omit scope to search all scopes (results follow precedence session → project → global → agent → federated).
Filter with scope= or memory_type= when you know where to look. Recall is for searching
beyond what was auto-injected (see below) — don't re-recall what's already in front of you.
Store — save anything worth remembering, immediately
Store the moment you learn something durable. Don't wait until the end of the session. Store conclusions, not transcript. Keep each memory to 1–2 sentences.
Store when you hit any of these:
- A correction — "No, we use DynamoDB here, not SQL." → store it so no agent makes that mistake again.
- A decided convention — "Every endpoint must have a pytest test before merge."
- A user preference — how they like work done, tools they prefer.
- A non-obvious project constraint — something you couldn't infer from the code.
memory_store(
content="Use DynamoDB for widgets-api; never SQL.",
scope="project",
memory_type="project",
key="widgets-database", # optional; auto-slugged from content if omitted
)Same key + scope upserts (updates in place) rather than duplicating.
Show full SKILL.md (313 more words)Show less
Share across all your projects — scope="federated"
When a lesson is durable and not specific to this repo — a reusable library gotcha, a
debugging trick, a tooling preference that holds everywhere — store it with
scope="federated" so it follows you into every project on this machine, not just this one.
memory_store(
content="tmux paste-buffer needs `-p` or multi-line input loses bracketed-paste framing.",
scope="federated",
memory_type="reference",
)Federated memories sit at the lowest recall precedence — a project-local fact with the
same key always wins — so federating is safe: it only adds a fallback, never overrides what's
true here. To un-share, memory_forget(key=..., scope="federated").
- Never federate secrets. Tokens, keys, and passwords are rejected automatically on a federated write — and they'd be exposed to every project anyway. Keep credentials out of memory entirely.
- When in doubt, use
project. Federate only what you're confident is reusable everywhere.
Forget — remove what's wrong or superseded
memory_forget(key="widgets-database", scope="project")Use this when a stored fact becomes outdated or was wrong. Prefer correcting (re-store with the same key) over leaving stale facts in memory.
For native memory, forgetting removes the CAO-managed topic file. For vault-backed memory,
it removes CAO's derived index entry and leaves the vault note in place for the operator to
edit or delete in Obsidian. The memory_forget result's action is authoritative; its
legacy deleted boolean means CAO completed the forget operation, not that the vault file
was removed. Inspect path when the action reports vault deindexing.
Auto-injection (already happening)
On launch, CAO writes the most relevant memories for this working directory into the file
your CLI reads on startup (Claude Code: .claude/CLAUDE.md; Codex: AGENTS.md; Kiro:
.kiro/steering/cao-memory.md). So you usually begin a task already knowing the project's
key facts — memory_recall is for digging up anything that wasn't injected.
Habits
- Recall before asking. The answer may already be stored.
- Store the instant you learn something durable — corrections, conventions, preferences, constraints.
- One fact per memory, 1–2 sentences. Conclusions, not conversation.
- Pick the right scope: user-wide →
global; this repo →project.
© awslabs, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in skills/cao-memory of awslabs/cli-agent-orchestrator.
Open the folder on GitHubat commit b29f40a
Compare with similar skills
Cao Memory next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Cao Memory this skillawslabs/cli-agent-orchestrator | 1.4k | — | ~1.4k | Automated safety check: Pass | Apache-2.0 | |
| Agent Squad Python Guide2FastLabs/agent-squad | 7.8k | — | ~4.7k | Automated safety check: Pass | Apache-2.0 | |
| Agent Squad for TypeScript2FastLabs/agent-squad | 7.8k | — | ~4.3k | Automated safety check: Pass | Apache-2.0 | |
| Codex with ChatGPT Planning LoopXiaoDuoYa/codex-with-chatgpt | 7.1k | — | ~11k | Automated safety check: Notes | MIT | |
| MemPalace Task HandoffMemPalace/mempalace | 59k | — | ~1.9k | Automated safety check: Pass | MIT | |
| agtx One-Shot Project Runnerfynnfluegge/agtx | 1.7k | — | ~3.8k | Automated safety check: Pass | Apache-2.0 |
Similar skills
Agent Squad for TypeScript
2FastLabs/agent-squad
Guide to building Node.js and TypeScript apps on the agent-squad package: orchestrator, agent types, classifier routing, storage, retrievers and MCP tools.
Codex with ChatGPT Planning Loop
XiaoDuoYa/codex-with-chatgpt
Uses ChatGPT in the browser as the planning and review brain for a Codex session, with Codex keeping all execution and ChatGPT reading the workspace through a bridge.
MemPalace Task Handoff
MemPalace/mempalace
Creates, hands off, claims, executes and closes agent tasks through the MemPalace logstream, with approval of the exact task before it is recorded.
agtx One-Shot Project Runner
fynnfluegge/agtx
Runs a whole project unattended on an agtx kanban board, decomposing the goal, starting tasks, unblocking workers and merging each result.
Puppetmaster Agent Orchestration
professorpalmer/Puppetmaster
Operates and supervises Puppetmaster, a multi-agent orchestrator, through its MCP tools or CLI, picking the right verb for edits, reviews, audits and long-running jobs.
More from awslabs/cli-agent-orchestrator
All 14 skills in this repo- Official
Cao MCP Apps
awslabs/cli-agent-orchestrator
Enable, operate, and extend CAO's MCP Apps surface — the host-rendered fleet dashboard visible inside MCP App hosts (Claude Desktop, ChatGPT, VS Code Copilot, Goose, Postman).
- Official
Agui Author
awslabs/cli-agent-orchestrator
Author live dashboard UI from an agent via the emitui MCP tool.
- Official
MCP Apps Builder
awslabs/cli-agent-orchestrator
Load the official MCP Apps builder skills (create-mcp-app, migrate-oai-app, add-app-to-server, convert-web-app) from github.com/modelcontextprotocol/ext-apps.
- Official
Cao Plugin
awslabs/cli-agent-orchestrator
Create a new CAO (CLI Agent Orchestrator) plugin. An agent skill from awslabs/cli-agent-orchestrator.
- Official
Cao Provider
awslabs/cli-agent-orchestrator
Create a new CLI agent provider for CAO (CLI Agent Orchestrator).
- Official
Cao Agent Routing
awslabs/cli-agent-orchestrator
Find and select the best installed CAO agent profile for a task before delegating with assign or handoff.
Related
Works with
Categories
Questions about Cao Memory
What does Cao Memory do?
Store, recall, and forget durable facts with CAO memory — user preferences, project conventions, decisions, and corrections that should persist across sessions and agents. Cao Memory is an agent skill from awslabs/cli-agent-orchestrator, published by the product's own GitHub organization. Store, recall, and forget durable facts with CAO memory — user preferences, project conventions, decisions, and corrections that should persist across sessions and agents.
When should I use Cao Memory?
Cao Memory fits situations like: tasks that involve Multi-agent orchestration.
How do I install Cao Memory in Claude Code?
Run `npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a claude-code`. Or copy the skill folder (skills/cao-memory in awslabs/cli-agent-orchestrator) into .claude/skills/cao-memory in your project. Claude Code loads it when a task matches its description.
How do I install Cao Memory in Codex?
Run `npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a codex`. Or copy the skill folder (skills/cao-memory in awslabs/cli-agent-orchestrator) into .agents/skills/cao-memory in your project. Codex loads it when a task matches its description.
Can I use Cao Memory in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add awslabs/cli-agent-orchestrator --skill cao-memory -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cao-memory, .gemini/skills/cao-memory, .github/skills/cao-memory and .opencode/skills/cao-memory in your project.
What does Cao Memory need to run?
SKILL.md names no scripts, command-line tools or credentials: Cao Memory is instructions for the agent only.
Does Cao Memory access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Cao Memory safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Cao Memory use?
Cao Memory is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Cao Memory use?
About 1.4k tokens (SKILL.md is roughly 5.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Cao Memory?
Skills that share tags, products or a category with Cao Memory: Agent Squad Python Guide (2FastLabs/agent-squad, 7.8k stars), Agent Squad for TypeScript (2FastLabs/agent-squad, 7.8k stars), Codex with ChatGPT Planning Loop (XiaoDuoYa/codex-with-chatgpt, 7.1k stars) and MemPalace Task Handoff (MemPalace/mempalace, 59k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Cao Memory?
awslabs (a GitHub organization, an official publisher) maintains it in awslabs/cli-agent-orchestrator, which has 1,396 GitHub stars. The repository holds 14 skills in this directory. The repository was last updated on October 8, 2026.
Source: awslabs/cli-agent-orchestrator on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.