MCP App Store
by KeenEthics
Thor icon

Thor MCP Server for ChatGPT

By Thor

Overview

Thor connects ChatGPT to the tools your team uses, infers projects, decisions, and commitments, and enriches your agents with up-to-date context. This gives ChatGPT a single MCP connection to efficiently pull the latest data from your team and keep everyone in sync. Use Thor to: • Help ChatGPT get the freshest picture of your team’s decisions before doing its work • Create high-quality issues, check for duplicates automatically, and link to supporting work • Extract action items, decisions, and commitments from meeting transcripts Thor can connect to: • Issues in GitHub and Linear • Team chats in Slack and Microsoft Teams • Meetings in Google Meet, Zoom, and Microsoft Teams • Documents in Notion

How to connect

  1. ChatGPT: Open the Thor listing in the ChatGPT app directory and choose Connect. ChatGPT listing Thor exposes 17 tools here.

Tools

add_standup_feedback

ChatGPT
Add items to your standup summary.

apply_work_item_feedback

ChatGPT
Persist the user's work-state: lifecycle updates, explicit status changes, and amendments when Thor's picture of their work is wrong. Call this whenever the user tells you how their work actually stands — progress, done, blocked, cancelled, a finished or reopened obligation, a blocker that is or is not real, a Definition of Done change, or that a prior Thor / Pulse / standup / reminder report was incorrect. Do not wait for the words "correct" or "override"; any actionable update to work items, obligations, blockers, or DoD belongs here. Actions: set_status, resolve_obligation, reopen_obligation, dismiss_blocker, assert_blocker, update_dod, clear_override. Without an active agent profile or Pulse / AGENT_PROFILE_AUTOMATION thread, each action requires that the work item is on the caller's plate (authored activity, owned meeting summary, or owned thread summary). Team-wide edits are allowed only in Pulse or profile context. Items that are not on your plate are denied per action. set_status requires status (todo, in_progress, blocked, done, or cancelled) and a note (reason). Never omit status or put it only in the note. Successful set_status results include linked_sources (empty if sources fail to load). If status_conflict is present, the override was recorded but an open/reopened linked issue/PR will keep the read model in_progress — do not tell the user the item is the requested status. Unresolved obligations or blockers can still keep the item open after reconcile even when status_conflict is absent. Prefer obligation_id / blocker_id / criterion_id from get_work_item_evidence digests.

create_issue

ChatGPT
Create a GitHub or Linear issue for the authenticated user's organization. Pass a human repository ref — prefer GitHub `owner/repo (e.g. sindresorhus/awesome) or a github.com URL, or a Linear team key / display name. Omit repo to use the caller's default repository memory when one exists. Prefer this over ask_thor for issue creation. If the ref cannot be resolved, call search_repositories with a looser query, then retry with one of those ref values (full owner/repo when the hit is GitHub). Title and description are required. The title is used as a strong hint; the shared issue_agent still formats the final title and body (including any repository issue template). By default this checks for similar open issues in the target repository. If any are returned, show them to the user and retry with the same arguments plus skip_duplicate_check=true` only after they confirm.

create_session

ChatGPT
Create a new session and return its identifier. Mints an unguessable uuid4, records an initial session owned by the current principal, and returns the id as a string. Store it and pass it back as a session_id argument on later calls to persist state across a session — only an id created this way resolves. State is keyed by the authenticated principal, so the id organizes sessions within a user; on an unauthenticated connection the id is the only thing standing between callers, which is why it is unguessable.

end_session

ChatGPT
End a session and delete all of its state. Validates the id like any other resolution (an unknown or foreign id is rejected), then deletes the session's key so the id no longer resolves.

get_answer

ChatGPT
Fetch the answer to a thor_troubleshooter or ask_thor (with agent_profile) run that returned "still running", or that your tool timeout cut: the run kept going and its answer is kept for a day. With session_id, the session's latest run; with run_id, that run (from the reply that started it); with neither, your own latest run, whichever tool started it. Waits up to wait_seconds (default 100) for the answer with a progress notification every 20 s, then returns it in the same shape the starting tool would have (Markdown, then a ``json block with session_id and run_id), or "still running" again, in which case call once more. Pass wait_seconds=0` to just check. Args: session_id: The session the run was started on. run_id: The run itself, when it was started without a session. wait_seconds: How long to wait in this call for the answer, 0 or more; default 100.

get_issue_context_pack

ChatGPT
Return Thor's cached issue context pack for a GitHub or Linear issue. The pack is assembled once and served on demand — ancestry, related work, repo docs, meetings, and decisions. Moat-only: the issue title/body are not re-sent (callers already have `repo + issue_number). A pack assembled in the last hour is served from cache. Pass refresh=true to regenerate. Optionally filter to named pack ## sections (e.g. ["Related work", "Decisions"]`).

get_meeting

ChatGPT
Return a structured record for one meeting: details, glossary-corrected transcript, decisions, and action items (including URLs for linked issues or PRs). Use this when the user asks for a specific meeting's transcript, attendees, decisions, or action items. Discover candidates with search_meetings first when the meeting_id is unknown. Prefer this over ask_thor for meeting lookup. Search by meeting_id (transcript UUID, or the calendar event UUID from search_meetings) or by title plus an optional YYYY-MM-DD date. Calendar events without a transcript still return details with an empty transcript.

get_standup

ChatGPT
Return a standup report for yourself or another user in one of the supported formats. ONLY use this tool when the user asks for actual standup report data formatted as json, markdown, Slack Block Kit, or a Teams Adaptive Card. For general work analysis, use get_work_item_evidence / search_work_item_evidence instead. Prefer send_standup_summary when the goal is to deliver a standup to a Slack/Teams thread.

get_work_item_evidence

ChatGPT
Load the evidence digest (findings, blockers, obligations, provider conflicts, Definition of Done) for specific work items. Prefer this (and search_work_item_evidence) for general work analysis. Use get_standup only when the user asks for an actual standup report in a specific output format.

list_my_work_items

ChatGPT
SCOPE: user (your linked plate). Todo list of work items you are linked to (meeting/thread ownership or authored issue/PR activity). Use for “my jobs / on my plate / what do I have outstanding.” Not organization-wide and not an agent-profile portfolio. For evidence filters or a named headline lookup, use search_work_item_evidence. Work items are linked to the user via meeting transcript ownership, thread summary ownership, or authored issue/PR activity.

search_meetings

ChatGPT
Discover meetings the caller is entitled to see. Returns only meeting_id, calendar_event_id, title, start_at, and end_at. Use this to find a meeting, then call get_meeting with meeting_id (or calendar_event_id as meeting_id) for the transcript (when one exists), attendees, decisions, and action items. Prefer this over ask_thor for meeting discovery. With no filters, returns the most recent meetings (up to 20), preferring meetings that already have a transcript over upcoming calendar-only events. Title or date filters rank by start time only so upcoming matches stay visible.

search_repositories

ChatGPT
Discover GitHub repositories and Linear teams the caller can file issues on. Prefer create_issue first when the user gave owner/repo, a Linear team key, or a unique team name. Call this only after create_issue could not resolve the target, or when the user is browsing ("which Linear teams do I have?"). Pass a returned `ref straight back to create_issue(repo=...). Prefer the full GitHub owner/repo` ref. Do not invent UUIDs.

search_work_item_evidence

ChatGPT
SCOPE: organization unless you pass scope=viewer. Evidence search (digest stubs: findings, coverage, DoD). Pass scope=viewer to search only YOUR linked items; scope=organization for org-wide recent activity. scope=profile requires an active agent profile and errors on this MCP tool. Use for blockers / ready-to-close / headline match. Do not use this to list your plate — that is list_my_work_items. Prefer this for exploratory work analysis. Follow up with get_work_item_evidence for details. Do not use get_standup for general "what did people work on" questions. Usage: - "Find the outreach videos item" -> query="outreach videos" (no finding_kinds) - "What's blocked?" -> finding_kinds=["active_blocker"] - "What looks ready to close?" -> finding_kinds=["provider_open_but_work_done"] - "Blockers on my plate" -> scope="viewer", finding_kinds=["active_blocker"] Do not pass finding_kinds on a named lookup. Kinds are a triage filter after you have a set, not part of finding a title. Viewer search without query/queries/finding_kinds is a worse list_my_work_items. status is the provider-backed read model. When a row has status_conflict, a human override disagrees — report the read-model status and explain the contradiction. Do not tell the user the item is the human status.

send_standup_summary

ChatGPT
Generate and post a standup summary as a thread reply in Slack or Teams. channel_external_id and thread_external_id are both required (MCP has no conversation context). Never use this to create a top-level channel message — post a root with send_message_raw (or equivalent) first, then pass that channel + thread id here.

thor_data_status

ChatGPT
Report what Thor is doing with the user's data, for each kind of data it reads: meetings, issues and repositories, work activity, work items, messages, and channels. Each one is "not_connected", "syncing" (pulling data from a connected tool), "processing" (Thor is reading or organizing what it pulled) or "ready". While a sync is open the sentence includes how many transcripts, issues, channels, and threads are in. Call it when the user asks what Thor is doing or whether their data is ready, or when another tool's result says it may be incomplete. Tell the user the message in plain words.

thor_onboarding

ChatGPT
Walk the user through connecting their tools to Thor, one step at a time. Call it with no arguments to start, or to start over. Every response names the next step to pass back, both for continuing and for skipping. What to ask next is not remembered between calls: each one re-reads what is actually connected, so call it again after the user says they have finished signing in rather than assuming it worked. Choosing an interest is also saved so the web app can resume; that save does not change what this tool asks next. With no arguments, ask which interest they want. Then follow that response. A connection step (step="meeting_provider" or step="project_management") is separate: it asks which tools they use and returns connect links. Do not gather every answer up front, and do not pass both tool arguments in the same call. Any step can be returned to at any time, including after the last one — if the user asks to go back, pass that step.

Capabilities

Writes

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/thor"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/thor)