# The tools personal assistants will reach for, and where they get stuck (slim) > Assistants such as Muse, ChatGPT and Claude are taking on errands. We reviewed 132 listings in 13 everyday categories. Most were built for people, and sign-in, payment and checkout are where agents stall. - Full: https://www.anchorterminal.com/blog/tools-for-personal-assistants.md (~5,250 tokens) · this version ~2,980 tokens · JSON https://www.anchorterminal.com/blog/tools-for-personal-assistants.json · canonical https://www.anchorterminal.com/blog/tools-for-personal-assistants - Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-14 - Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on By Quynh Tran-Thanh, 14 October 2026. 2,071 words, about 9 min to read. Assistants such as Claude, ChatGPT and Meta's Muse now connect to outside services. Today most of those tools were made for a person. On 14 October we added seven categories of everyday tools to the directory [[2]](#ref-2). - 132: listings in 13 everyday categories - 20: of them have no route an outside agent can use - 1: of 55 in the seven new categories is agent-ready Directory data, 14 October 2026. Source [3]. ## The errand categories score lowest We grade every listing out of 100 against the same checklist, and the average across the directory is 57.7 [[3]](#ref-3). Average score by category, out of 100 | Category | Average score | | --- | --- | | Email | 64 | | Messaging | 64 | | Maps | 62 | | Scheduling | 59 | | Contacts | 57 | | Health data | 57 | | Bank data | 54 | | Travel | 49 | | Shopping | 48 | | Restaurant and local bookings | 35 | | Subscriptions and personal finance | 23 | | Home services | 22 | Each series has a zero-based scale. The pattern is in who the vendor sells to. ## How an outside agent gets in For each listing we record whether a developer can sign up and get a key, whether access goes through a partnership or a sales call, or whether there is no route for an outside agent at all. How an outside agent can get access, listings per category | Category | Self-serve | Partner or sales | Not for agents | | --- | --- | --- | --- | | Email | 14 | 0 | 0 | | Messaging | 14 | 0 | 0 | | Bank data | 8 | 7 | 0 | | Health data | 7 | 5 | 0 | | Travel | 6 | 5 | 0 | | Shopping | 5 | 0 | 0 | | Home services | 2 | 1 | 7 | | Restaurant and local bookings | 1 | 4 | 4 | | Subscriptions and personal finance | 1 | 0 | 9 | Each series has a zero-based scale. Shopping looks open in that chart, but none of the five shopping listings lets an agent place an order through a self-serve API. MCP servers follow the same line. Listings with an MCP server | Category | Listings | With an MCP server | | --- | --- | --- | | Messaging | 14 | 11 | | Email | 14 | 10 | | Maps | 13 | 9 | | Health data | 12 | 5 | | Travel | 11 | 4 | | Scheduling | 10 | 5 | | Restaurant and local bookings | 9 | 3 | | Home services | 10 | 2 | | Shopping | 5 | 1 | | Subscriptions and personal finance | 10 | 0 | Each series has a zero-based scale. ## The categories, one by one **Restaurant and local bookings.** [Square Appointments](/tools/square-appointments) is the one self-serve booking API in the category, and it acts for the venue, not the customer [[8]](#ref-8). **Home services.** Seven of ten listings are not for agents, among them [Angi](/tools/angi), whose terms forbid access by robots and other automatic devices [[11]](#ref-11). **Subscriptions and personal finance.** Nine of ten are apps with no API an outside agent can call. **Shopping.** Catalogue search works. **Travel.** [Duffel](/tools/duffel) lets a developer sign up and book flights in test mode with no card or contract [[17]](#ref-17). **Health data, contacts and desktop automation.** These lean towards developers and score near the directory average [[3]](#ref-3). The assistants themselves are graded too. ## Where agents stall The same few steps stop an agent across every category. ### Signing in A consumer app assumes a person will make the account and prove who they are. Signing in to a budgeting app built for people 1. **Find the app** (agent can do it). The assistant can find Rocket Money or Emma and read what it does [[14]](#ref-14) [[24]](#ref-24) 2. **Make an account** (needs the person). Both start with an account a person makes in the app or on the web [[14]](#ref-14) [[24]](#ref-24) 3. **Connect the bank** (needs the person). Bank accounts connect through Plaid, Akoya or Open Banking, with the person signing in to the bank [[14]](#ref-14) [[24]](#ref-24) 4. **Read the budget** (blocked). There is no public API, and both sets of terms forbid bots and scrapers [[14]](#ref-14) [[24]](#ref-24) YNAB shows the other route. Signing in where the app has an API 1. **Sign up** (needs the person). A person signs up in a browser [[15]](#ref-15) 2. **Approve the assistant** (needs the person). The person creates a token or approves an OAuth app, and can choose a read-only scope [[15]](#ref-15) 3. **Read and edit the budget** (agent can do it). The agent calls the REST API with the token [[15]](#ref-15) 4. **Stay within the limit** (costs tokens). Each token gets 200 requests an hour, and a 429 response doesn't say how long to wait [[15]](#ref-15) Some services built for agents shorten even the first step. ### Booking Booking a table shows how much of a flow can be ready for agents and still be closed to a new assistant. Booking a table through OpenTable 1. **Get access** (blocked). An assistant needs approval as an OpenTable partner and an agreement first [[9]](#ref-9) 2. **Search availability** (agent can do it). Availability search covers a window of up to 720 minutes [[9]](#ref-9) 3. **Check the policy** (agent can do it). Policy endpoints return deposits, card holds and cancellation cut-offs before booking [[9]](#ref-9) 4. **Ask the person** (needs the person). Any deposit, hold or service fee should go to the diner before the agent commits [[9]](#ref-9) 5. **Lock the slot** (costs tokens). The lock expires within 5 minutes, so a slow confirmation means searching again [[9]](#ref-9) 6. **Pay the hold** (needs the person). Card details are tokenised with Spreedly or Stripe, from the diner's card [[9]](#ref-9) 7. **Make the booking** (agent can do it). A unique X-Request-Id stops a retried booking being made twice [[9]](#ref-9) ### Checkout and payment Buying something is where every route we read hands back to a person. Buying groceries through Instacart 1. **Turn the list into a cart** (agent can do it). Two MCP tools turn a recipe or list into an Instacart page [[16]](#ref-16) 2. **Choose products and prices** (blocked). There is no product search or price endpoint [[16]](#ref-16) 3. **Pick a store and pay** (needs the person). The person picks a store and checks out on Instacart [[16]](#ref-16) 4. **Track the order** (blocked). There is no order or order status endpoint [[16]](#ref-16) Checkatrade's booking flow ends the same way, at a payment link the person completes [[13]](#ref-13). Paying through an agent payment protocol 1. **Set the limits** (needs the person). Under AP2 the person signs a mandate in advance, bound to the agent's key [[26]](#ref-26) 2. **Get the price** (agent can do it). Under x402 the server answers with status 402 and what the request costs [[27]](#ref-27) 3. **Pay** (agent can do it). The agent signs a stablecoin transfer from its wallet, or uses a token from the person's saved card under ACP [[27]](#ref-27) [[28]](#ref-28) 4. **Get the result** (agent can do it). The agent sends the request again with proof of payment [[27]](#ref-27) The protocols exist, but they are early. ### Cancelling Cancelling a subscription has no API route among the listings we read. Cancelling a subscription for someone 1. **Find the charge** (agent can do it). A budgeting API such as YNAB's lists a person's transactions [[15]](#ref-15) 2. **Reach the merchant** (blocked). The finance apps we list cancel only inside their own apps [[14]](#ref-14) 3. **Call or write instead** (costs tokens). An errand service such as Asmi calls, texts or emails the business as a paid task [[1]](#ref-1) 4. **Allow the outreach** (needs the person). The person grants the scope that lets Asmi contact people on their behalf [[1]](#ref-1) ## Where the tokens go When there is no API, an assistant can still open a browser or drive a desktop app. Tools that do have an MCP server can cost tokens another way. MCP tools a server exposes | Server | Tools | | --- | --- | | Courier | 170 | | MailerSend | 127 | | LiteAPI | 111 | | Redox | 107 | | Cal.com | 63 | | AgentMail | 36 | | Checkatrade | 2 | | Instacart | 2 | Each series has a zero-based scale. AgentMail loads all 38 of its hosted tools at once [[25]](#ref-25), and LiteAPI's hosted MCP exposes 111 generated tools [[29]](#ref-29). ## What would help Most of this is in reach of the companies that already have APIs. - A key or an OAuth app a developer can get without a sales call, as Square, YNAB and Duffel allow [[8]](#ref-8) [[15]](#ref-15) [[17]](#ref-17). - A read-only scope, so a person can let an assistant look without letting it act, as YNAB has [[15]](#ref-15). - An idempotency key on every booking or payment, as OpenTable's X-Request-Id and Square's idempotency keys are [[9]](#ref-9) [[8]](#ref-8). - Policies and prices returned before the commitment, as OpenTable's policy endpoints do [[9]](#ref-9). - A payment step an agent can finish within limits the person set, or at least a payment link it can hand over, as Checkatrade returns [[13]](#ref-13). - A clear rule for agents. Amazon's Agent Terms require agents to identify themselves, which at least says where the line is [[4]](#ref-4). The apps we tagged "not for agents" are graded on the same checklist as everything else, so the day one of them opens a route, its score will show it. ## How we counted Counts and averages come from the directory as built on 14 October 2026. A category's average is the mean score of its graded listings. --- - Older: [Cypherpunks write code, and now agents read it](https://www.anchorterminal.com/blog/rea-cypherpunks-write-code.md) - Newer: [How to make your MCP server discoverable](https://www.anchorterminal.com/blog/how-to-make-your-mcp-server-discoverable.md) - All posts: https://www.anchorterminal.com/blog/index.md ## References Listings read from the directory on 14 October 2026. Each listing carries the vendor sources it was graded from. [1] Anchor Terminal, Asmi listing. [2] Anchor Terminal, changelog, 14 October 2026. Source for the seven categories added that day. [3] Anchor Terminal, tool rankings and category pages, 14 October 2026. Source for listing counts, average scores, access routes, MCP counts, tool counts, x402 counts and "not for agents" tags. [4] Anchor Terminal, Amazon Creators API listing. [5] Anchor Terminal, Walmart Affiliate Marketing API listing. [6] Anchor Terminal, Best Buy Developer API listing. [7] Anchor Terminal, Etsy Open API v3 listing. [8] Anchor Terminal, Square Appointments listing. [9] Anchor Terminal, OpenTable APIs listing. [10] Anchor Terminal, Resy listing. [11] Anchor Terminal, Angi listing. [12] Anchor Terminal, Taskrabbit Partner API listing. [13] Anchor Terminal, Checkatrade listing. [14] Anchor Terminal, Rocket Money listing. [15] Anchor Terminal, YNAB listing. [16] Anchor Terminal, Instacart Developer Platform listing. [17] Anchor Terminal, Duffel listing. [18] Anchor Terminal, Booking.com Demand API listing. [19] Anchor Terminal, Kiwi.com MCP Server listing. [20] Anchor Terminal, Terra API listing. [21] Anchor Terminal, Google People API listing. [22] Anchor Terminal, UiPath listing. [23] Anchor Terminal, personal assistants category, 14 October 2026. Source for the category's listings and average score. [24] Anchor Terminal, Emma listing. [25] Anchor Terminal, AgentMail listing. [26] Anchor Terminal, Agent Payments Protocol (AP2) listing. [27] Anchor Terminal, x402 at Anchor Terminal. [28] Anchor Terminal, Agentic Commerce Protocol (ACP) listing. [29] Anchor Terminal, LiteAPI listing.