ConnectOnion is an open-source, template-first toolkit for building, debugging, deploying, and operating AI agents. Developers use its command-line tools and Python runtime to create agents, add tools, connect services, deploy them, and make them callable by other agents, while the catalogue entries are related agents, skills, and instructions.
Read one batch of authorized session messages or mail and write the extraction notes the rem-maintain runner will organize — every durable fact with who said it, when, and its source id. Read the supplied material file and write the complete notes to the requested output file.
Maintain an AI-owned co rem from explicitly authorized source messages through co ai and ordinary Markdown file operations. Use for organizing, correcting, and consolidating that notebook, not for changing source apps or installing learned Skills.
What an organisation's page in the notebook is made of — the fixed sections, when a company earns a page at all instead of staying a field on a person's, and the rules that keep it from becoming a second copy of everyone who works there. Composed into every stage that writes one.
Where the user's own words are inside Codex CLI sessions, how that store is laid out, and what this source has already taught us. Loaded alongside whichever stage Skill is running when the material comes from Codex.
What is true of WhatsApp chats as a co rem source — who is speaking, why the agent's own replies are missing, how a group differs from a person, and what a photo in the batch means. Loaded alongside whichever stage Skill is running when the material comes from WhatsApp.
Retrieve context from an existing co rem through the local co rem inspection commands. Use to answer questions about its recorded people, projects, decisions, knowledge or commitments; this preview does not start collection or edit notes.
Work with Canny customer feedback through co canny — a weekly feedback digest, the top-voted open requests, finding duplicates, and marking a request planned while telling its voters. Use when the user mentions Canny, feature requests, feedback boards, votes, roadmap status or the Canny changelog.
Work with the user's Linear issues through co linear — list their open issues for a standup, file a bug from a report, comment on an issue and close it. Use when the user mentions Linear, a Linear issue id like ENG-123, "my tickets", "file this bug", or closing an issue.
Read and answer Slack with co slack — what a channel said, a thread, a search across the workspace, and a reply in the right thread. Use for "what did.
Hand the work discussed in this session (every message the user wrote, your replies quoted, where the code is) to another person, by email or straight to their ConnectOnion agent, after removing secrets and getting the user's approval of the exact text; and receive one from a pasted prompt that carries a coh1., coh2.…
Think one page through slowly over several turns in one thread — ask many questions, pick the five worth pursuing, study each, and write only when a question yields an insight.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: