Vellum Assistant is a personal AI assistant that remembers information about users, learns their preferences, and takes actions across connected apps. It is intended for people who want an assistant that can manage conversations, unfinished work, and proactive notifications over time. The catalogue skills, hooks, instruction, and setting configure or extend how the assistant works.
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 vellum-ai/vellum-assistant --skill api-mappinggit clone --depth 1 https://github.com/vellum-ai/vellum-assistantWrote 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/vellum-ai/vellum-assistant/api-mapping)<a href="https://agentmods.dev/skills/vellum-ai/vellum-assistant/api-mapping"><img src="https://agentmods.dev/badge/skills/vellum-ai/vellum-assistant/api-mapping.svg" alt="Measured on agentmods" 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 | $0.00012 | $0.00891 |
| Opus 5 | $0.00006 | $0.00445 |
| Sonnet 5 | $0.00002 | $0.00178 |
| Haiku 4.5 | $0.00001 | $0.00089 |
Grade A, and why
api-mapping 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 3d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You can record and analyze the API surface of any web service using the map CLI.
CLI Setup
IMPORTANT: Always use host_bash (not bash) for all map commands. The API mapping CLI needs host access for Chrome CDP, network recording, and browser automation - none of which are available inside the sandbox.
map is a CLI tool that should be available on your PATH. Do NOT search for the binary or try to discover how the CLI works. Just run the commands as documented below.
Typical Flow
When the user wants to map a web service's API (e.g. "Map the Notion API" or "Figure out how Figma's API works"):
-
Ask about the objective - Ask the user: "What service do you want to map?" and "What are you trying to build or integrate with?" This helps focus the recording on the relevant parts of the API surface.
-
Choose mode - Ask the user: "Should I browse automatically, or do you want to drive?" This determines whether to use auto mode or manual mode:
- Auto mode: The CLI launches a headless browser, navigates the service, and records API calls automatically. Best for broad discovery.
- Manual mode: A Chrome window opens for the user to interact with the service while the CLI records all API traffic in the background. Best for capturing a specific workflow.
-
Run the mapping - Execute the appropriate
mapcommand:- Auto mode:
map <domain> --json - Manual mode:
map <domain> --manual --json - For longer sessions:
map <domain> --duration 120 --json
- Auto mode:
-
Wait for recording to complete - In auto mode, the CLI will browse and record for the default duration (60 seconds) then stop. In manual mode, the CLI blocks until the user closes the browser or presses Ctrl+C. The command outputs a JSON summary of all discovered endpoints.
-
Analyze the API map - Review the output and present findings to the user:
- List discovered endpoints grouped by resource type (e.g.,
/api/v1/users,/api/v1/documents) - Note authentication patterns (Bearer tokens, cookies, API keys)
- Identify CRUD operations and their HTTP methods
- Highlight any WebSocket or streaming endpoints
- Call out rate limiting headers or pagination patterns
- List discovered endpoints grouped by resource type (e.g.,
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago Changed · -17 lines 5ea882903680
- 7d ago First seen · 84 lines · 12 tokens per session scan A a17d8f8dd661
api-mapping is a skill published in the GitHub repository vellum-ai/vellum-assistant (1,194 stars, last pushed today), licensed MIT. It adds 12 tokens to every session and 891 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
web-deserialization
Insecure deserialization to RCE for web apps. Use when the app deserializes attacker-controlled data - cookies/tokens/hidden fields/params that are serialized blobs, VIEWSTATE, Java/PHP/.NET/Python/Node apps. Triggers - base64 starting rO0AB or hex AC ED 00 05 (Java), PHP serialize O-prefix, VIEWSTATE, python pickle…
ms365-connect
Connect to Microsoft 365 (OneDrive, Outlook mail, Calendar, Teams) via Microsoft Graph using the open-source python-o365 (O365) library. Use when the user wants to list/download OneDrive files, read or send Outlook email, list calendar events, or post a message to a Teams channel from a Sutando workflow.
calling-llms
Use when sending chat completions through liter-llm and routing to a specific provider via the provider/model prefix. Covers the chat call shape, provider routing, modelhint, message roles, and error categories.
running-the-proxy
Use when running the liter-llm api OpenAI-compatible gateway — virtual keys, per-key rate limits, budgets, cost tracking, and model routing. Covers the TOML config and the 22 REST endpoints.
streaming-responses
Use when streaming tokens incrementally from an LLM via liter-llm over SSE or async iterators. Covers chatstream, delta handling, and null-content chunks.
kafka-event-driven-design
Kafka event-driven architecture designer and reviewer, at the application/client layer. ALWAYS use when designing, reviewing, or troubleshooting how a service produces or consumes Kafka events — topic and partition-key design, producer and consumer client configuration, consumer group topology, event schema definition…