By use case
These are the workflows agents actually run against this twin, each naming the MCP tools and REST surfaces involved. Anything named here that is not fully modelled says so in place.Post and edit messages in a channel or thread
Post a message, reply in a thread, edit it, or pull it back. MCP:slack_send_message, slack_schedule_message. REST: chat.postMessage, chat.update, chat.delete, chat.scheduleMessage, chat.deleteScheduledMessage.
React to and pin a message
Add or remove a reaction, read who reacted, and pin the message that matters. MCP:slack_add_reaction, slack_get_reactions. REST: reactions.add, reactions.remove, reactions.get, pins.add, pins.remove, pins.list.
Read channel and thread history
Pull a channel’s backlog or a thread’s replies before summarizing or acting on it. MCP:slack_read_channel, slack_read_thread. REST: conversations.history, conversations.replies.
Manage channel membership
Create a channel, invite or remove members, and archive it when the work is done. MCP:slack_create_conversation, slack_search_channels, slack_list_channel_members. REST: conversations.create, conversations.invite, conversations.join, conversations.leave, conversations.kick, conversations.members, conversations.archive, conversations.setTopic, conversations.setPurpose.
Look up users
Resolve a mention or an email to a user, and read their profile. MCP:slack_search_users, slack_read_user_profile. REST: users.list, users.info, users.lookupByEmail, users.profile.get, users.profile.set.
Search messages
Find a message across the workspace instead of walking channel history. MCP:slack_search_public, slack_search_public_and_private. REST: search.messages.
Share files, bookmarks, and canvases
Upload a file, bookmark a link on a channel, or create a canvas. MCP:slack_create_canvas, slack_read_canvas, slack_update_canvas, slack_read_file. REST: files.upload, files.info, files.list, files.delete, bookmarks.add, bookmarks.remove, bookmarks.list, canvases.create, canvases.edit, canvases.delete. slack_create_canvas, slack_read_canvas, slack_update_canvas, slack_read_file, files.upload, files.info, files.list, files.delete, canvases.create, canvases.edit, canvases.delete are shape only — see What you can rely on before you grade on them.
What you can rely on
Each surface below carries two rulings that are deliberately kept apart: heat is how deep it should be, ruled per milestone; state is how deep it is today, measured by the twin’s own tests. Neither moves because the other did. A surface is in exactly one state, and the five below are different answers to “can I write a task against this”:@pome-sh/twin-slack declares 18 MCP tools and 50 REST surfaces, read from the twin’s own fidelity.inventory.json (last updated 2026-08-10):
Shape only
Read these; do not grade on them. The response shape is right and the behaviour behind it is not asserted, so a criterion that checks a value here can pass for the wrong reason.Out of scope
Real Slack OAuth, SCIM, admin APIs, and Enterprise Grid are not modelled — the twin issues static bearer tokens, and an agent that needs a real OAuth flow should run against Slack directly. These surfaces are named rather than left to the catch-all, so the 501 is documented and test-backed:Known divergence, ruled
25 divergences are on the ledger for Slack: measured against the real API, found to differ, reviewed, and accepted. Each is registered and reverse-tested, so one that upstream heals becomes a signal rather than a surprise. The numbers are stable identifiers, not positions — a retired divergence leaves its number behind, so a gap in this list is a divergence that closed.Vendor-only tool
The vendor declares it and the twin answersunknown_tool. An agent doing the right thing is marked down for it.
Where it runs. Both doors serve the same twin:
pome twin start slack boots it as a local process, and pome sandbox create --twin slack starts a hosted sandbox running the same image. The states above are properties of the twin, so they hold on either.
Everything above is what the twin declares. What it was measured to do is a different number: the comparison against real Slack runs daily and publishes, surface by surface, how many matched, which drifted, and which are ruled exceptions with the reason written out in full — at status.pome.sh/twin/slack, which also states when it last ran and how old the captured baseline it compares against is. No count is copied onto this page, because copying one is how it goes stale.
Read from FIDELITY.md and fidelity.inventory.json in pome-sh/digital-twins, at the commit .github/twins-ref pins (2e41939).
Quickstart
Point your agent at it
For interactive development without a full task run, start the standalone twin:POME_SLACK_REST_URL is the Web API root. POME_SLACK_MCP_URL is the MCP transport endpoint. POME_AUTH_TOKEN is a short-lived JWT that authenticates every request on the sandbox.
A local pome run injects the same three into your agent’s process, pointed at that run’s own sandbox rather than standalone.
For MCP-based agents:
Task seed shape
Slack task seeds are flat, withteam, users, channels, files and emoji. A channel carries its own members and messages.
Generate it; do not copy it. The block below is what the twin starts with, printed by the CLI from the twin’s own declared state — so it parses against the twin you are about to seed, today and after the twin changes:
twins: ["slack"] in the task’s ## Config block and name the file <task>.seed.json, beside the task’s .md. That ## Config is the only place the twin gets named: the file above is flat, so it carries no twin id of its own.
Seed file
A task is not the only way in.pome twin start and pome sandbox create take that very same file — no wrapper, no second shape, nothing to convert — and each takes the twin’s name beside it, since the file cannot supply one:
pome twin new-seed slack github writes it:
twin start and sandbox create want the name as well, and why a task’s sidecar does not — the ## Config beside it already named the twin. From two twins up the file is a per-twin envelope { <twin>: <seed> }, which does name its twins, though twin start still has to be told which one of them to boot. Replace, not merge, applies wherever the file lands.
Build your own world is that job end to end — generating the file, editing it down, both doors, and how to tell whether every field you wrote actually landed.
Example tasks
Ready-made examples you can run or copy to see the twin in action:Catalog
Run one
pome-suggest-tasks and pome-run-task skills, installed by the graded capstone.