Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ap311036/ews-meeting-mcp --skill ews-meeting-mcpgit clone --depth 1 https://github.com/ap311036/ews-meeting-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ap311036/ews-meeting-mcp/ews-meeting-mcp)<a href="https://agentmods.dev/skills/ap311036/ews-meeting-mcp/ews-meeting-mcp"><img src="https://agentmods.dev/badge/skills/ap311036/ews-meeting-mcp/ews-meeting-mcp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ap311036/ews-meeting-mcp/ews-meeting-mcp"><img src="https://agentmods.dev/badge/skills/ap311036/ews-meeting-mcp/ews-meeting-mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00070 | $0.02981 |
| Opus 5 | $0.00035 | $0.01491 |
| Sonnet 5 | $0.00014 | $0.00596 |
| Haiku 4.5 | $0.00007 | $0.00298 |
Grade A, and why
ews-meeting-mcp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
EWS Meeting Agent
Use the ews-meeting-mcp MCP tools to read company Outlook mail, send self-only Jira reports, and schedule meetings through an on-prem Exchange/EWS calendar.
Workflow
- For company Outlook mail, call
ews_setup_check, thenews_search_emailswith the narrowest known filters. Present the returned metadata, use the exact selectedidand optionalchangekeywithews_read_email, and summarize the returned message. These tools are read-only: they do not mark messages as read, move messages, delete messages, or change labels. - At the start of a scheduling session, call
ews_setup_check. If it returnsready: false, show itsuser_messageas-is, or showsetup_commandverbatim in a fencedbashblock, and stop. Do not ask the user to paste a password into chat or MCP config; do not ask for attendee emails or continue with EWS tools until the user says they fixed setup. Then callews_setup_checkagain before continuing. If the user asks how to set up an Outlook-style meeting signature, callews_signature_setup_guideand show the returnedsample_html,recommended_path, andsetup_command; this helper does not require EWS credentials. - For the Thursday Jira weekly-report automation, after it assembles the previous and current report, call
ews_send_email_to_selfwithsubject,body,body_format, and a stable, non-secretidempotency_key. It has no recipient,to,cc, orbccargument; it sends only to the configuredEWS_EMAILthrough EWSCreateItemfollowed bySendItem. Never substitute Gmail orto: me. Reuse the same idempotency key for retries; treatduplicate_confirmation,confirmation_in_progress, and unknown delivery as stop conditions until Sent Items and the audit log are inspected. - Parse the user's request into attendees, candidate rooms, date range, duration, subject, body, and location.
For Taiwan/local office scheduling, pass datetime arguments with an explicit local offset such as
+08:00.Zis accepted by the tool but means UTC, not local time. - If any attendee is not a complete email address, call
ews_resolve_attendeesfirst. Do not ask the user to provide full attendee email addresses before trying company directory resolution. - Use exactly resolved emails for scheduling. If a name has zero matches or multiple matches, ask the user to clarify or choose from the returned candidates before continuing. Prefer the host's interactive multiple-choice UI or clickable choice controls when available; otherwise show a short numbered list of candidates. For ambiguous matches, show candidate names and emails as choices; do not ask the user to type the full email manually.
- If the user did not mention rooms, call
ews_list_roomswith the attendee count if known, then ask whether a meeting room is needed using the returnedoptions. Prefer the host's interactive multiple-choice UI or clickable choice controls when available; include "no meeting room" and "auto-pick any room with enough capacity" choices. The defaultsource: "auto"tries Exchange room-list discovery and falls back to configured policy rooms; offer the user a choice among the returned roomvalues or "no specific room"; do not rely on a hand-written room list. - If the user wants a room but does not choose a specific room, call
ews_suggest_slotswithrequire_room: trueand norooms; the tool uses dynamic Exchange rooms when available, then falls back to configured rooms, and filters rooms with knowncapacitybelow the attendee count.Pmeans persons. - If the user chooses specific rooms, pass the selected room
values or emails inroomswhen callingews_suggest_slots. - Show suggested slots with their
available_rooms. When asking the user to pick a slot or room, prefer the host's interactive multiple-choice UI or clickable choice controls when available; otherwise show a short numbered list. Omitworkday_start,workday_end, andavoidto use local policy defaults. - When the user picks a slot and room, call
ews_create_meeting_previewwith the selected room inrooms. - For recurring meeting requests, pass a structured
recurrenceobject to preview and confirmed create calls. Weekly Monday/Wednesday uses{"type":"weekly","interval":1,"weekdays":["MO","WE"],"range":{"type":"numbered","count":10}}. Every business day until 2026-07-26 uses{"type":"weekly","interval":1,"weekdays":["MO","TU","WE","TH","FR"],"range":{"type":"end_date","end_date":"2026-07-26"}}. Business days mean Monday through Friday only. If the user gives weekdays but no end date, occurrence count, or explicit no-end choice, ask for one before previewing. - Show the exact preview: attendees, rooms, start, end, subject, body, location,
signaturestatus, recurrence if present, andconfirmation_id. - Only after the user explicitly confirms, call
ews_create_meeting_confirmedwithconfirm: trueand the exactconfirmation_idreturned by preview. - After a successful confirmed create, call
ews_verify_meetingwith the returned itemidandchangekeywhen available, then report the organizer item status and roomresponse_status. Treatunknownroom response as pending/unknown, not as failure.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 79 lines · 70 tokens per session scan A 95eb8b9ee12d
ews-meeting-mcp is a skill published in the GitHub repository ap311036/ews-meeting-mcp (3 stars, last pushed 16d ago), licensed MIT. It adds 70 tokens to every session and 2,981 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.