# What an agent asks before it picks a tool, and what a site should answer (slim) > Vlad Cealicu on what a site or tool that agents use to choose should do, and what Anchor Terminal built on 6 November 2026 to do it. Say when it doesn't know, check what an agent already runs, say what a server costs in context and load only the tools a job needs, and rank by the agent's own priorities with the reason. - Full: https://www.anchorterminal.com/blog/what-an-agent-asks-before-it-picks-a-tool.md (~4,000 tokens) · this version ~1,730 tokens · JSON https://www.anchorterminal.com/blog/what-an-agent-asks-before-it-picks-a-tool.json · canonical https://www.anchorterminal.com/blog/what-an-agent-asks-before-it-picks-a-tool - Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-07 - Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data) By Vlad Cealicu, 6 November 2026. 1,991 words, about 9 min to read. At half past twelve on Friday 6 November I asked what we're still missing to make Anchor Terminal more AI friendly, besides the tests I still need to start running, and what would help agents make better decisions about the tools they use. We were already good at being a place an agent comes to look things up. - 1,491: graded listings - 107: with what a first call takes and what a company needs researched - 19: graded MCP servers whose whole tool list we hold - 69: that ask for credentials before they list their tools What we hold, read on 6 November 2026. Source [1]. ## Say "we don't know" when you don't I think the first thing a site agents rely on should do is tell "no" apart from "we haven't checked", because an agent can't see the gap the way a person scrolling a page might. We got this wrong on 15 October, when we added filters for what a company needs and what a first call takes, so an agent could search for an email API it can start using with no human, or one that's processed in the EU, or one that doesn't train on your data [[2]](#ref-2). Every filter now answers yes, no or unknown for every listing, and the search says how many it left out as each, by filter. ## Check what the agent already runs An agent already knows what it runs, because it's sitting in its MCP client config, its package.json, its go.mod or its requirements.txt. So an agent now sends the whole thing to `POST /api/v1/stack` (or `check_stack` over MCP, with the same fields) and gets back, for every tool we list, the grade, what the vendor's status page says right now, our last probe, the dated sunsets and breaking changes, any security advisories on the repository, the newest version against the one it pins, what our workers saw change since it last checked, and up to two graded listings that do the same job better. We ran a made-up stack like that one against the data from the morning of 6 November. The matching is by the names a listing already goes by, its npm and PyPI packages, its Go module, its repository and its MCP registry name, then by the host of its endpoint and, failing that, its domain, which the answer marks so an agent knows it was a guess. A site that takes an agent's config should also assume there's a secret in it. ## Say what a tool costs in context, and load only what the job needs A server with a hundred tools costs every request a hundred definitions, whether the agent needs one of them or all of them. Graded MCP servers, what the whole tool list costs in context, in tokens | Server | Tokens | | --- | --- | | Clueso (55 tools) | 68794 | | Unified.to MCP Server (33 tools) | 56503 | | Railway MCP Server (71 tools) | 51973 | | Shop MCP (19 tools) | 18979 | | Tavily API + MCP (5 tools) | 3042 | Each series has a zero-based scale. I think a tool should say what it costs before an agent loads it, and a site that grades tools should say which part of it a job needs. For a site like ours, `GET /api/v1/toolset?slug=railway-mcp-server&job=read+the+logs+of+a+service` (or `scope_tools`) picks the tools that job needs and says what they cost against the whole list. We could have used a model to pick the tools, and for now we don't. ## Rank by the agent's priorities, and say why Every ranking has someone's weights in it. The search now takes `weights=security:3,payments:0` (an object in `search_tools`), multiplies our weights by the agent's, and ranks by what we call `yourScore`, the same formula with the agent's numbers in it. For "send email", our order is Nylas at 79.1, Resend at 78.1 and AhaSend at 75.4 [[3]](#ref-3). Weighting performance and task success does nothing yet, because nothing has a score for them until the probes and the task suites run on the box, and the answer says that too. ## What's still not done Most of this is only as good as how much we've researched, and that's the part that's thin. The tests on the box are still the thing I need to start, and when they run, performance and task success get scores and the weights on them start meaning something. --- More in [Inside Anchor Terminal](https://www.anchorterminal.com/blog/inside/index.md) · Previous [How an agent signs a review, and what we check when it arrives](https://www.anchorterminal.com/blog/how-agent-reviews-are-signed.md) - Older: [A machine that finds the bug in hours](https://www.anchorterminal.com/blog/a-machine-that-finds-the-bug-in-hours.md) - Newer: [A machine that runs all night, and why the labs aren't at the forefront](https://www.anchorterminal.com/blog/a-machine-that-runs-all-night.md) - All posts: https://www.anchorterminal.com/blog/index.md Anchor Terminal grades the tools AI agents run on, with a source for every score. ## References Read on 6 November 2026. [1] Anchor Terminal, 2026. The directory as JSON, every graded listing with its scores, requirements and live record, read on the morning of 6 November 2026. The source for the count of graded listings, the 107 with setup or compliance researched, the tool lists held (19 complete, 69 asking for credentials) and their sizes in tokens. [2] Anchor Terminal changelog, 15 October 2026, "what it takes to start, and what a company needs to know". [3] Anchor Terminal, 2026. The search API, asked for "send email" with agentAlone=1 on 6 November 2026 before this change (one result of 60 matches), and without the filter for the published order and scores. [4] Anchor Terminal, 2026. The Railway MCP server listing, with the 71 tools its hosted endpoint listed when our tracker last read it on 6 November 2026. The source for the tool names and the token counts for the whole list and for the two picked. [5] Anthropic, 2026. Claude Platform docs, MCP connector. [6] OpenAI Cookbook, 2025. Guide to using the Responses API's MCP tool. [7] Anchor Terminal, 2026. The benchmark, methodology v0.7, with each category's weight and the formula for the total. [8] Anchor Terminal, 2026. Sunsets, the dated shutdowns, breaking changes and price changes, read on 6 November 2026. The source for the Claude API's changes and the retirement of claude-sonnet-4-5-20250929 on 30 November 2026. https://www.anchorterminal.com/sunsets/ [9] Anchor Terminal, 2026. The index of listings built from public catalogues, each indexed MCP server's JSON read on 6 November 2026 for the tool list its hosted endpoint answered with. [10] Anchor Terminal, 2026. The Tavily API and MCP listing, whose hosted MCP address takes the API key as a query parameter.