Agent skill · Audited from source

Outlook

skill · vellum-ai/vellum-assistantMIT licenceAudited 14 Oct 2026 · source 19bc86343b63

★ 1,425pushed 1 day ago
!CAUTIONGrade C · 79/100 · 7 findings · 19bc863
1,425repo stars · whole vellum-ai/vellum-assistant

Is it safe? With care — CAUTION, grade C (79/100). We read this skill's source at commit 19bc86343b63 and found no filesystem, network or shell access and no hard-coded secrets. The audit raised 7 findings, 1 critical or high. What it does: Manage Outlook email — drafting, sending, organizing, rules, vacation replies, and inbox analysis

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Outlook (vellum-ai/vellum-assistant)”. Retrieved 14 Oct 2026. https://modelsatlas.com/skills/vellum-ai/outlook

Trust audit

Audit 0.4.1 · source read 14 Oct 2026

Trust score79/100!CAUTION C
0 critical1 high0 medium5 low1 info

CAUTION means the audit found things worth knowing before you trust this skill. Grade C is shared by 84 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS5
L1 Static sourceSecrets, path traversal, supply chain, hidden codePASS0
L2 InstructionsPrompt injection, authority framing, hidden directivesFAIL2
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: instruction override, no behavioural pass.

Findings (7)

HIGHPrompt injectionprompt.hide_from_user · SKILL.md:132
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
- Do not tell the user a draft exists until the tool or script returns success with a `draftId`. If creation fails or is interrupted, say so and offer the email copy in chat (clipboard or an in-app no
Why
asks the agent to act without the user's knowledge
Fix
—
INFOPrompt injectionprompt.conditional_escalation · SKILL.md:135
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
**To actually send:** use `bun run scripts/outlook-email.ts send-draft` with the draft ID after the user has reviewed it. Only run send-draft when the user explicitly says "send it" or equivalent.
Why
No further detail recorded for this rule.
Fix
—

Audit 0.4.1 · source sha 19bc86343b63 · 14 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publishervellum-ai GitHub organizationOwner of the repository this skill is read from
  • Source foldervellum-ai/vellum-assistant/skills/outlookThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdict19bc86343b63A new audit runs when the source changes
  • Repository statusNot archivedLast push 1 day ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Manage Outlook email — drafting, sending, organizing, rules, vacation replies, and inbox analysis

---
name: outlook
description: Manage Outlook email — drafting, sending, organizing, rules, vacation replies, and inbox analysis
compatibility: "Designed for Vellum personal assistants"
metadata:
  icon: assets/icon.svg
  emoji: "📧"
  vellum:
    category: "email"
    display-name: "Outlook"
    user-invocable: true
---

This skill provides Outlook-specific operations beyond the shared **messaging** skill. For cross-platform messaging (send, read, search, reply), use the messaging skill. Outlook operations depend on the messaging skill's provider infrastructure - load messaging first if Outlook is not yet connected.

## Script Reference

All operations use CLI scripts that return JSON:

- **Success**: `{ "ok": true, "data": ... }`
- **Failure**: `{ "ok": false, "error": "..." }`

| Script              | Operation       | Description                                                              |
| ------------------- | --------------- | ------------------------------------------------------------------------ |
| `outlook-email.ts`  | `draft`         | Create email drafts in the Drafts folder (including reply drafts)        |
| `outlook-email.ts`  | `send-draft`    | Send an existing draft (**requires explicit user confirmation**)         |
| `outlook-email.ts`  | `forward`       | Create forward drafts, preserving attachments                            |
| `outlook-email.ts`  | `trash`         | Move messages to Deleted Items                                           |
| `outlook-manage.ts` | `categories`    | Manage message categories (add, remove, list available)                  |
| `outlook-manage.ts` | `follow-up`     | Track messages with Outlook's native flag system                         |
| `outlook-manage.ts` | `attachments`   | List and download email attachments                                      |
| `outlook-manage.ts` | `rules`         | Create, list, and delete server-side inbox message rules                 |
| `outlook-manage.ts` | `vacation`      | Get, enable, or disable auto-reply (out-of-office) settings              |
| `outlook-manage.ts` | `unsubscribe`   | Unsubscribe from mailing lists (**requires explicit user confirmation**) |
| `outlook-scan.ts`   | `sender-digest` | Scan inbox and group messages by sender for declutter workflows          |
| `outlook-scan.ts`   | `outreach-scan` | Identify cold outreach senders (no List-Unsubscribe header)              |

### Email Operations

```bash
# Draft an email (--to is optional; omit it to create a draft with no recipient)
bun run scripts/outlook-email.ts draft --to "[email protected]" --subject "Hello" --body "Message body"
bun run scripts/outlook-email.ts draft --subject "Hello" --body "Message body"

# Send an existing draft (REQUIRES user confirmation before execution)
bun run scripts/outlook-email.ts send-draft --draft-id "AAMk..."

# Forward a message
bun run scripts/outlook-email.ts forward --message-id "AAMk..." --to "[email protected]" --comment "FYI"

# Trash a message
bun run scripts/outlook-email.ts trash --message-id "AAMk..."
```

### Management Operations

```bash
# Categories - add categories to a message
bun run scripts/outlook-manage.ts categories --action add --message-id "AAMk..." --categories "Blue Category,Important"

# Follow-up - flag a message for follow-up
bun run scripts/outlook-manage.ts follow-up --action track --message-id "AAMk..."

# Attachments - list attachments on a message
bun run scripts/outlook-manage.ts attachments --action list --message-id "AAMk..."

# Rules - create a server-side inbox rule
bun run scripts/outlook-manage.ts rules --action create --name "Archive newsletters" --conditions '{"senderContains":["newsletter"]}' --actions '{"moveToFolder":"Archive"}'

# Vacation - enable auto-reply
bun run scripts/outlook-manage.ts vacation --action enable --internal-message "I'm out of office" --external-message "I'm currently away"

# Unsubscribe from a mailing list (REQUIRES user confirmation before execution)
bun run scripts/outlook-manage.ts unsubscribe --message-id "AAMk..."
```

### Scan Operations

```bash
# Sender digest - scan inbox grouped by sender
bun run scripts/outlook-scan.ts sender-digest [--query "..."] [--time-range "90d"] [--account "..."]

# Outreach scan - identify cold outreach senders
bun run scripts/outlook-scan.ts outreach-scan [--time-range "90d"] [--account "..."]
```

Scan scripts store message IDs in the daemon's in-memory cache (via `assistant cache`) and return a lightweight summary with a `cache_key`. This keeps thousands of message IDs out of the conversation context. To retrieve cached message IDs for a specific sender:

```bash
assistant cache get <cache_key> --json
# Returns: { "ok": true, "data": { "[email protected]": ["msgId1", "msgId2", ...], ... } }
```

## Email Routing Priority

When the user mentions "email" - sending, reading, checking, decluttering, drafting, or anything else - **always default to the user's own email (Outlook)** unless they explicitly ask about the assistant's own email address (e.g., "set up your email", "send from your address", "check your inbox"). The vast majority of email requests are about the user's Outlook, not the assistant's @vellum.me address.

Do not offer the assistant's own email as an option unless the user specifically asks. If Outlook is not connected, guide them through Outlook setup.

## Connection Setup

### Outlook

1. **Check connection health first.** Run `assistant oauth ping outlook`. This checks whether the user's Outlook/Microsoft account is connected and the token is valid.
2. **If no connection is found or the ping fails:** Load the `vellum-oauth-integrations` skill. The skill will evaluate whether managed or your-own mode is appropriate and guide the user accordingly.

## Error Recovery

When an Outlook script fails with a token or authorization error:

1. **Try to reconnect silently.** Run `assistant oauth ping outlook`. This often resolves expired tokens automatically.
2. **If rec

Excerpt, first 6,001 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 19bc863

Download folder ↗
git clone https://github.com/vellum-ai/vellum-assistant
git -C vellum-assistant checkout 19bc86343b63

The author documents no install command, so this uses the repository directly. Our verdict covers commit 19bc86343b63 only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. !CAUTION C 79 / 100

    first audit · source 19bc86343b63

README badge

For maintainers · shows the latest audit

ModelsAtlas: CAUTION CMODELSATLASCAUTION · C
[![ModelsAtlas: CAUTION C](https://modelsatlas.com/skills/vellum-ai/outlook/badge.svg)](https://modelsatlas.com/skills/vellum-ai/outlook/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Add this badge to your README

[![ModelsAtlas: CAUTION C](https://modelsatlas.com/skills/vellum-ai/outlook/badge.svg)](https://modelsatlas.com/skills/vellum-ai/outlook/)

Questions about Outlook

What does the Outlook skill do?
An AI Assistant that’s easy to setup, does your work 24/7, knows your preferences and gets better over time.
Is Outlook safe to install?
With care. The audit graded it C (79/100) and found 7 things worth knowing before you trust this skill, listed below with the exact line each was found on.
What can Outlook access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (19bc86343b63), read on 2026-10-14. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.