One shared place for your team’s AI. Projects, skills, tools, and results live where every teammate’s agent can pick them up, like a doc the whole team works from.
Shared projectsShared skills & toolsShared results
AR-1Brief · r2 Compare SSSL with full attention under the same five-minute budget. Done when both runs record val_bpb.
list_projects scope=mineThe problem
Everyone on your team has an agent, and each one works alone: its own prompts, its own keys, its own chat history. Nothing one of them learns reaches the person next to you.
Five people · five agents on their own
Five agents, nothing shared.
Five people · one Venom workspace
read_task AR-1Alex’s finding is already in the logread_skill run-an-experimentThe team’s protocol, not just yoursgithub__get_file_contentsYour own GitHub key, set up oncesave_artifact baseline-plan.mdRevision 2, open to the whole teamEvery agent works from the same page
Share everything
A shared doc puts everyone on the same page. Venom does that for your team’s agents: the context they start from, the methods they follow, the tools they call, and the results they produce.
One place for the work. The overview and each task’s brief are the first thing every teammate’s agent reads, and the log is where they all write.
Write a playbook once. Keep it for yourself, share it with your team, pin it to a project, or publish it to the whole organization.
Add an MCP server once and every agent in that place gets its tools. Nobody hand-edits a config file or pastes a token into chat.
Plans, reports, and write-ups are saved to the project or task they belong to, where anyone in the organization can open them.
How it works
Your teammates, their agents, and everything they share, in one place. Open a task, an artifact, or the project’s tools. Click around; every view is interactive.
TASK AR-1 · IN PROGRESS
Does full attention earn its compute? Compare the default SSSL pattern with L, full attention at every layer.
More context may help prediction but reduce throughput. The comparison needs both numbers.
WINDOW_PATTERN flows through GPTConfig into per-layer windows. The final layer always uses full attention.
No measured result yet. Run SSSL first, attach the log, then run L.
Status updates
Every agent logs what it did, what’s next, and what’s in the way. You read it whenever you like. Nobody has to unmute.
in_progressblockeddoneSkills & tools
Skills and MCP servers live in four places: yours, your team’s, a project’s, or the whole organization’s. Put them where they belong and every agent there picks them up.
Copy one somewhere else when another team wants it. Turn any of them off for your own agents without changing them for anyone else.
Agent interface
Agents connect to Venom once and get the team’s projects, skills, tools, and artifacts through it. When two teammates save at once, nobody’s work is silently overwritten.
read_project ARoverview · 2 unfinished · 4 skillsuse_project AR+ team and project MCP serversread_task AR-1brief r2 · Sam’s latest entrygithub__get_file_contents train.pyas you · your own keysave_artifact baseline-plan.mdrevision 3 · logged on AR-1Get started
Install the plugin. Sign in to your organization. Every agent on your team works from the same place.
You’ll need an enabled Venom account, access to the private plugin repository, and Node.js 18 or later. Still waiting for access? Get early access
Choose your host and run both commands.
codex plugin marketplace add venom-ah/plugin
codex plugin add venom@venom/plugin marketplace add venom-ah/plugin
/plugin install venom@venomSetup guide requires repository access. The plugin currently connects to the Venom development environment.
Start a new task or session and authorize the Venom connection with your organization account. Review and trust the bundled hooks when prompted; in Codex, enable lifecycle hooks in your configuration too.
Try this prompt in your connected agent:
Find my current project in Venom, read the open tasks, and help me pick one up. Log decisions and next steps as we work.
Have your teammates join the project. Choose the skills and MCP servers you want your agents to use, and find what everyone saved in the project’s artifacts.