> ## Documentation Index
> Fetch the complete documentation index at: https://docs.poddata.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# The MCP server

> PodData's own MCP server: its address, how to add it to Claude Code with a key or by signing in, and every tool it has.

PodData's MCP server gives an assistant PodData's tools: every read route has one, and so does each change a diligence needs. It's at:

```
https://api.poddata.ai/mcp
```

It speaks Streamable HTTP: one JSON answer to each `POST`, no sessions. Each tool calls the API's own route as you, so it keeps the same checks, limits, counts and error messages, and answers twice: as text an assistant can read, and as the route's JSON.

> [!NOTE]
>
> This is PodData's server, with PodData's data. A docs site's own "add to your assistant" button adds a server that searches these pages; it isn't this one.

## Add it with a key

1. Keep the key in the environment as `PODDATA_API_KEY`.

2. In the project's folder, run this as it is, in single quotes, so Claude Code reads the key from the environment each time it starts and never writes it down:

   ```bash theme={null}
   claude mcp add --transport http --scope project poddata https://api.poddata.ai/mcp --header 'Authorization: Bearer ${PODDATA_API_KEY}'
   ```

3. Start Claude Code again. `/mcp` lists `poddata` with its tools.

Any other client that speaks MCP over Streamable HTTP takes the same address and the same header.

## Add it by signing in

With no key, the server asks the client to sign in, and the person signs in to PodData and picks their team.

1. Add it without a header:

   ```bash theme={null}
   claude mcp add --transport http --scope project poddata https://api.poddata.ai/mcp
   ```

2. In Claude Code, run `/mcp`, pick `poddata` and sign in. PodData's sign-in opens in the browser.

3. The sign-in asks for two scopes: `quotes:read`, for every tool that reads, and `trackers:write`, for every tool that changes something, such as confirming a link or saving a report's page. Without `trackers:write`, those tools aren't listed.

A sign-in works for an hour and renews itself for 30 days. Only a team whose plan includes the API can sign in.

## When a tool is refused

A refusal comes back as the tool's answer, marked as an error, with the API's own sentence: a limit, a company that isn't a target yet, a link confirmed for another company. Read it and do what it says. A call past the key's requests a minute waits, as the API's does: see [Keys and limits](https://docs.poddata.ai/keys-and-limits.md).

## Its tools

106 tools, in the order the server lists them. A tool that changes something needs the `trackers:write` scope when you sign in; with a key, every tool is there.

| Tool | What it does | Kind | Reads |
| - | - | - | - |
| `search_quotes` | Find what people said on podcasts and in long interviews: exact quotes with the speaker, the show, the date, the second and a page on poddata.ai to cite. | Reads | `GET /quotes` |
| `search_counts` | How many quotes, passages companies wrote, people, shows and companies one search finds, each counted as its own list would list it, before reading any of them. | Reads | `GET /search/counts` |
| `search_places` | Which places search\_passages can search here, its `in`: filings and releases (documents), decks (slides), call transcripts (transcripts) and news (articles), each with whether it holds anything to search yet. | Reads | `GET /search/places` |
| `list_trackers` | The team's trackers: what each follows, whether it's running, and when it last found a quote. | Reads | `GET /trackers`, `GET /trackers/{id}` |
| `preview_tracker` | What a tracker would have caught in the last 90 days, before making it. | Reads | |
| `create_tracker` | Make a tracker that sends each new matching quote by email. | Changes something | |
| `tracker_matches` | The quotes a tracker found, newest first, each with why it matched and its match id (for not\_this). | Reads | `GET /trackers/{id}/matches` |
| `not_this` | Mark a match as wrong for its tracker, so the tracker learns from it. | Changes something | |
| `list_matches` | Every one of the team's trackers' matches in one list, newest first: quotes, passages companies wrote for trackers that follow documents, and posts from the accounts PodData follows for trackers that follow posts, each with why it matched, its tracker and its match id. | Reads | `GET /matches` |
| `update_tracker` | Change one of the team's trackers: rename it, pause or resume it, change how it emails (each quote, a daily digest, a weekly digest, or off), or who on the team gets it. | Changes something | |
| `delete_tracker` | Delete one of the team's trackers. | Changes something | |
| `match_feedback` | Tell a tracker a match was good, or not this, or take either back (clear), so the tracker learns what the team wants. | Changes something | |
| `list_saved` | The quotes the person signed in has saved, and the passages they picked from transcripts and saved, newest first, each with its exact words and how to cite it, and how many there are. | Reads | `GET /saved` |
| `save_quote` | Save a quote for the person signed in, to find it again in the workspace's Saved. | Changes something | |
| `unsave_quote` | Take a quote out of the person's saved quotes. | Changes something | |
| `save_passage` | Save up to 80 words the person picked from one turn of a transcript they're shown, to find again in the workspace's Saved. | Changes something | |
| `unsave_passage` | Take a passage the person saved out of Saved, by its id, sp\_…. | Changes something | |
| `list_saved_filters` | The sets of filters the person signed in has saved by name for the workspace's Search and All matches, newest first, each with its list and its filters. | Reads | `GET /filters` |
| `get_usage` | The team's counts by day and kind: calls through the API and assistants (api), quotes served through them (api\_quote), sources read whole (whole), searches, previews, matches and deliveries, the last 30 days unless from and to say otherwise. | Reads | `GET /usage` |
| `find_company` | Look a company up by name, by ticker (NVDA, or NASDAQ:NVDA), or by its SEC CIK, a FIGI or an ISIN. | Reads | `GET /companies`, `GET /companies/{id}` |
| `list_events` | A company's own events: earnings calls, investor and analyst days, conference sessions, shareholder meetings and the like, with the date (and whether it's announced or estimated), the fiscal period, the webcast, and what was filed with each. | Reads | `GET /events`, `GET /events/{id}` |
| `read_transcript` | A company's own call or event, turn by turn: each speaker named with their role and company (or marked unconfirmed), when each spoke, the chapters (the prepared remarks, the Q\&A and each analyst's question, with who asked), a link to the company's replay at each turn, and each turn's words once our speaker checks have passed. | Reads | `GET /transcripts/{id}`, `GET /transcripts/{id}/chapters` |
| `list_documents` | What a company filed with the SEC or published: annual and quarterly reports, earnings and press releases, proxy and registration statements, and slide decks, newest first, each with the day filed and a link to the original. | Reads | `GET /documents` |
| `search_passages` | Find what companies wrote in their filings, releases and decks: passages of at most 80 words, best first, each with the document, the page, the section, a link to that page of the original and the passage's own id. | Reads | `GET /passages` |
| `search_slides` | Find slides in companies' decks, such as how others present a metric: each slide once, with its title, its best words (at most 80), the company, the event and a link to the slide. | Reads | `GET /slides` |
| `get_summary` | PodData's summary of a company's event (the headline, what management said, the outlook and capital allocation), of a filing, release or deck, of a call's transcript, or of a podcast episode (the headline, what a speaker argued, what one expects, what was said of a company): one line, short (the default) or long, every line checked against and linked to the page, slide or second it rests on. | Reads | `GET /events/{id}/summary`, `GET /documents/{id}/summary`, `GET /transcripts/{id}/summary`, `GET /episodes/{id}/summary` |
| `get_figures` | What a company reported in its 10-Qs and 10-Ks, from the filings' own tagged data: revenue, operating income and every line of the income statement, balance sheet and cash flow, and its segments, newest period first, each value with its filing on EDGAR. | Reads | `GET /companies/{id}/figures` |
| `read_document` | A filing, release, deck or an outlet's article, by its id from list\_documents: the record, its outline (sections with their pages), and its pages from from\_page, each with what's on it, a link to the original at that page, and its words where PodData opens that kind of document to you (otherwise quote it with search\_passages). | Reads | `GET /documents/{id}`, `GET /documents/{id}/pages`, `GET /documents/{id}/text`, `GET /passages/{id}/source` |
| `slide_history` | The same slide in the company's earlier decks, newest first, by a slide's page id from search\_slides or read\_document: each version's deck, date and link, and the labels and numbers it added, changed or dropped since the version before it. | Reads | `GET /slides/{id}/history` |
| `list_transcripts` | Transcripts of companies' own calls and events, newest event first, each with its speakers named (or marked unconfirmed) and their role on the day. | Reads | `GET /transcripts` |
| `list_recordings` | Recordings of companies' own events, newest first: each with its length, where the Q\&A starts, the company's own replay, and its transcript's id. | Reads | `GET /audio`, `GET /audio/{id}` |
| `live_events` | Companies' events coming up live, live now or over, each with its state, when it went live, the company's webcast and where the Q\&A began. | Reads | `GET /live`, `GET /live/{id}` |
| `live_transcript` | A live call from a position: its sections, interruptions and end as they happen, and its words where PodData opens them to you. | Reads | `GET /live/{id}/transcript` |
| `company_timeline` | A company's events, reports and decks, and quotes, counted by month (or week or year): for events, how many we hold the transcript of, only the replay of, or only the date; the quotes its people said and the quotes that name it. | Reads | `GET /companies/{id}/timeline` |
| `company_people` | A company's people: those quoted while at the company, each with how many quotes and the day of the latest, most quoted first; then the officers its SEC filings name, each with the filing's title and a link to it. | Reads | `GET /companies/{id}/people` |
| `list_types` | The kinds of company event (earnings\_call, investor\_day and the rest) and of document (each with its SEC form, its kind and its group), with the ids list\_events, list\_documents and list\_transcripts take. | Reads | `GET /event-types`, `GET /document-types` |
| `look_up_person` | Find every place a person has spoken in public in their work: podcasts, interviews, talks, company events, hearings and TV. | Changes something | `GET /lookups/{id}` |
| `person_sources` | Every source PodData has for a person, one per conversation, newest first, each with where it stands: quoted, transcribed, queued, found only, or can't be fetched. | Reads | `GET /people/{id}/sources` |
| `find_person` | Look people up in PodData's directory by part of a name or an alias, narrowed by kind (ceo, founder, investor, researcher and so on) or company, A to Z, the most quoted first or the most recently quoted first. | Reads | `GET /people` |
| `find_name` | Find who or what a name means in PodData's directory: people, companies and shows whose name, other name or ticker starts with what you give, up to eight, best first. | Reads | `GET /directory` |
| `get_person` | One person from PodData's directory, by name or id: their role, company and page, and their latest ten confirmed quotes, newest episode first, each with its exact words and a page to cite. | Reads | `GET /people/{id}` |
| `person_timeline` | How often a person has spoken in public, counted by month (or week or year): each period's conversations, where they stand (quoted, transcribed, queued, found only, can't be fetched), the quotes we have, and the conversations by others that name them, with totals. | Reads | `GET /people/{id}/timeline` |
| `named_by_others` | Other people's conversations that name a person, newest first: each episode with how many times someone else said their name, the first second anyone did with a link to the show's own recording there, the confirmed speakers who did, and the quotes from it that name them. | Reads | `GET /people/{id}/named` |
| `list_lookups` | The team's person lookups this month, newest first, each with how it ended, and how many of the plan's lookups a month it has used. | Reads | `GET /lookups` |
| `list_shows` | Every show PodData follows, by name, each with its id, how we follow it, how many feeds it has, its latest episode's date, and how many quotes it has in how many of its episodes. | Reads | `GET /shows` |
| `get_show` | One show PodData follows, by its id or name: how we follow it, how many feeds it has, its latest episode's date, how many quotes it has in how many of its episodes, its own site and its feed's language. | Reads | `GET /shows/{id}` |
| `list_episodes` | The episodes PodData holds, newest first: what a show put out in a week, which episodes a person was in, or which ones name a company. | Reads | `GET /episodes` |
| `get_episode` | One episode, by its id from list\_episodes or a quote: who spoke and how we know, its chapters, every quote under its chapter with its exact words and a page to cite, the companies and people it names with counts, and its transcript turn by turn, each turn with its speaker, its second and a link there. | Reads | `GET /episodes/{id}` |
| `get_quote` | One quote by its id: its exact words, who said it (and their role then), the show, the episode, the date and the second, the page to cite, a link to the recording at that second, and what it names and claims. | Reads | `GET /quotes/{id}` |
| `quote_source` | Where a quote came from, in one call: the show's or the company's own recording from the quote's second, the episode or event, its chapters with whose titles they are, who spoke when with a link to each turn, and the other quotes near it in time order. | Reads | `GET /quotes/{id}/source` |
| `earlier_quotes` | What a quote's speaker said on the same topic before: up to five of their quotes from earlier episodes, oldest first, found by how close in meaning each is to this one. | Reads | `GET /quotes/{id}/earlier` |
| `get_coverage` | What PodData covers and how fresh it is: the feeds it follows, the people with a confirmed quote, the episodes it indexed in the last 7 and 30 days, and the median minutes from publication to indexed. | Reads | `GET /coverage`, `GET /coverage/people` |
| `list_lists` | The team's own lists of people and companies, each with how many it holds and how many trackers follow it, and the plan's limits. | Reads | `GET /lists`, `GET /lists/{id}` |
| `create_list` | Make one of the team's lists of people and companies, with a name and, if you have them, its first people and companies by directory id (find\_person and find\_company give ids). | Changes something | |
| `update_list` | Change one of the team's lists: rename it, add people and companies by directory id, or take them off. | Changes something | |
| `delete_list` | Delete one of the team's lists. | Changes something | |
| `match_names` | Match names against PodData's directory for one of the team's lists, a name on each line with a company or ticker after a comma if known (or a CSV with name, company and ticker columns). | Reads | |
| `list_targets` | The companies the team looks at for diligence, newest first, each with how many of its links are confirmed. | Reads | `GET /targets` |
| `add_target` | Add a company to the team's targets: by its PodData id; for a company that files with the SEC, by its ticker or SEC number, and PodData reads its last year of filings from EDGAR within the hour, as many companies a day as the plan allows; or, for a company PodData doesn't hold, by its name and its own domain, which goes in as a proposed link for PodData's team to confirm. | Changes something | |
| `find_filer` | Look up a company that files with the SEC by its ticker (HIMS, NYSE:HIMS) or its SEC number, in SEC's own list of filers. | Reads | `GET /filers/{filer}` |
| `remove_target` | Take a company off the team's targets. | Changes something | |
| `company_links` | How PodData finds a company in each source: its domain, Google place IDs, product IDs, accounts, SEC and provider numbers, registrations and licences, each proposed, confirmed or struck, with the page that shows it's theirs. | Reads | `GET /companies/{id}/links` |
| `company_records` | Records from agencies' and companies' own files, APIs and pages tied to a company by a confirmed link, newest first: breaches, recalls, warning letters, inspections, cases, deals, providers, awards, registrations, trademarks, licences and the like, each with its source, its day, at most 80 words of its text and its link. | Reads | `GET /companies/{id}/records` |
| `record_revisions` | One record, and each earlier version of it, oldest first, with when PodData read it saying so. | Reads | `GET /records/{id}`, `GET /records/{id}/revisions` |
| `company_signals` | Reviews, stories, community and social posts, ads, videos, search-interest points, web-traffic estimates and product listings tied to a company by a confirmed link, newest first, each with its link, its day, a rating or counts, at most 80 words, and how it was read: when, in what order, by what method, and the platform's own total and average. | Reads | `GET /companies/{id}/signals` |
| `get_signal` | One review, post, ad, video or search-interest point by its id: its link, its day, a rating or counts, at most 80 words, and how it was read: when, in what order, by what method, and the platform's own total and average. | Reads | `GET /signals/{id}` |
| `company_licensed` | Figures read about a company through your team's own connections to data vendors, such as its funding rounds from your team's PitchBook, newest first, each marked Your licence with the vendor and the day it was read. | Reads | `GET /companies/{id}/licensed` |
| `propose_link` | Propose how PodData finds a target in one source: its own domain, a Google place ID, a product's ID on a retailer, a review profile, a brand's own account, an SEC, provider or federal number, a state registration or licence, or the exact legal name a registry writes with its state. | Changes something | |
| `propose_rival` | Propose a company as one of a target's rivals, read beside it on a comparison. | Changes something | |
| `confirm_link` | Confirm one of a target's links, so PodData's readers read against it, or one of its rivals. | Changes something | |
| `strike_link` | Strike one of a target's links that isn't theirs, such as another company's listing of the same name, or a rival that isn't one, such as a supplier. | Changes something | |
| `company_rivals` | The companies proposed as a company's rivals, each proposed, confirmed or struck, with the page that shows they compete. | Reads | `GET /companies/{id}/rivals` |
| `target_coverage` | For one of the team's targets, whether the read now last asked for (read\_target) is going, done or held by a day's cap, for it and each confirmed rival, with what's still to run: wait while it's going, come back after midnight UTC or go on when it's held, go on when it's done. | Reads | `GET /targets/{company}/coverage` |
| `read_target` | Ask PodData to read one of the team's targets now, once its links are confirmed: each reader that reads against a confirmed link of the company reads it first on its next run, within the quarter hour, ahead of its usual cadence and under every source's daily cap, which a reader at its cap waits out. | Changes something | |
| `find_links` | Read one of the team's targets' own site, at its proposed or confirmed domain, and propose the accounts its pages link to (Instagram, TikTok, X, YouTube, Facebook, LinkedIn), each with the page that links to it, and list the addresses on its location pages for the person to match to listings. | Changes something | |
| `find_rivals` | Find one of the team's targets' rivals and test each like for like: candidates from a list you give (a customer's list of rivals is a proposal), from the records PodData holds, from stories and deals that name others beside the target, and from the open web where PodData allows it. | Changes something | |
| `target_folder` | Everything PodData holds about one of the team's targets, in one place: for the target, each confirmed rival and its listed parent and comparables, source by source, how many records, signals and documents each holds, their first and last days, and the address of each as a CSV file and as JSON rows; its quotes' two files; the team's own licensed figures and diligence reports on it; and every source that holds nothing for it, with where it stands and why. | Reads | `GET /targets/{company}/folder` |
| `numbers_rating` | A company's reviews from one source over a window, counted on PodData's server: the average, the count, the share at each star and the share of one and two stars, each with a 95% interval, and the platform's own all-time total, average and histogram from its latest pull. | Reads | `GET /companies/{id}/numbers/rating` |
| `numbers_rating_by_link` | The same figures for each of a company's confirmed links on one source, such as each Google place of a clinic network or each product on a retailer, weakest average first, each with its count, so the weakest can be listed. | Reads | `GET /companies/{id}/numbers/rating-by-link` |
| `numbers_rating_rivals` | A company's rating on one source against each of its confirmed rivals over the same window: a row a company, the rivals' reviews pooled, and the gap. | Reads | `GET /companies/{id}/numbers/rating-rivals` |
| `numbers_rating_windows` | A company's rating on one source in the last few months against the months before (three against nine unless asked), with the change, for the company and each confirmed rival. | Reads | `GET /companies/{id}/numbers/rating-windows` |
| `numbers_complaints` | Counts what a company's low-rated reviews on one source complain about, against each confirmed rival's: you give the topics, each a label and the terms that find it, and PodData counts how many unhappy reviews with text use each, their share, and how many places or products they came through, with up to five samples a topic to check that the terms found what they meant. | Reads | `GET /companies/{id}/numbers/complaints` |
| `numbers_series` | One field of a source's records a period at a time, such as a casino's monthly adjusted gross receipts from the Illinois Gaming Board (source igb-reports, field total\_agr), for the last 12 months against the 12 before unless asked, for the company and each confirmed rival that has the record, with totals, the change and the company's share of the group. | Reads | `GET /companies/{id}/numbers/series` |
| `numbers_timeline` | A company's dated events, newest first: its breaches, recalls, warning letters, awards, licences, cases and deals from agencies' and parties' own records, and the news stories about it, each with its own day, its source and its link. | Reads | `GET /companies/{id}/numbers/timeline` |
| `numbers_ads` | Counts a company's ads in Meta's US library against each confirmed rival's, on PodData's server: the ads running on one day (on, today if left out) and on an earlier one (vs, four weeks before), the change, what started and what stopped between them, and the running ones by platform. | Reads | `GET /companies/{id}/numbers/ads` |
| `numbers_ads_weeks` | The ads running at each of a company's weekly pulls of Meta's US library over a span (weeks to to, 12 weeks to today if left out), with what started and what stopped since the pull before, for the company and each confirmed rival: a row a company's pull, for a chart of ads over time. | Reads | `GET /companies/{id}/numbers/ads-weeks` |
| `company_ads` | A company's ads in Meta's US library, from PodData's weekly pulls of its confirmed Facebook page: the days it has pulls, its counts on a day (on, today if left out) and an earlier one (vs, four weeks before), and its ads, each with its start day (Meta's own), where it runs, at most 80 of its words, its page in the library and where it stands. | Reads | `GET /companies/{id}/ads` |
| `get_ad` | One ad from a brand's page in Meta's US library, by its id from company\_ads or an exhibit's evidence: at most 80 of its words, where it runs, its start day, whether it's still running or when it was last seen, the pulls it was seen in, and its page in the library, where Meta shows it. | Reads | `GET /ads/{id}` |
| `company_site` | The pages of a company's own website that PodData reads weekly for a team's target and its confirmed rivals (its home, products, pricing, customers, about, leadership, locations, careers, news, security and terms pages, as each site has them), on a day (on, today if left out): each page's address, its title, the day it was first and last read, the day its words last changed, the route that got it, and every version. | Reads | `GET /companies/{id}/site` |
| `site_page` | One page of a company's own website by its address (url, from company\_site), as it was on a day (on, today if left out): its words, whole where PodData may show them and at most 80 otherwise, its versions, and, given an earlier day (vs), the passages added and removed between the two versions. | Reads | `GET /companies/{id}/site/page` |
| `numbers_site` | Puts a company beside each confirmed rival on what their own websites say, counted on PodData's server from the pages it reads weekly, as each was last read on a day (on, today if left out): pages read, kinds of page found, the products and services each says it sells, the plans it prices (and whether its pricing page shows prices), the customers and partners it names, its locations, the leaders its pages name, its open roles and the certifications it says it holds. | Reads | `GET /companies/{id}/numbers/site` |
| `numbers_site_changes` | What changed on a company's own website and each confirmed rival's between two days (vs, four weeks before on if left out, and on, today): pages first read, pages whose words changed, pages taken down, and what their pages newly say and no longer say. | Reads | `GET /companies/{id}/numbers/site-changes` |
| `list_reports` | The team's diligence reports on its targets, newest change first, each with its status and version. | Reads | `GET /diligence-reports` |
| `start_report` | Start a diligence report on a company: an overview and five empty pages (customer, competitive, news, trends, expert), which you then write a page at a time with save\_report\_page. | Changes something | |
| `get_report` | A diligence report filled: every exhibit's number tool asked now, each figure in its place, each cited quote or review at most 80 words, and what the check refuses, where it is and why. | Reads | `GET /diligence-reports/{id}` |
| `save_report_page` | Save one page of a diligence report, or its overview, over the version you read (rev). | Changes something | |
| `check_report` | Run the check on a diligence report as it reads now, and mark it checked when nothing is refused. | Changes something | |
| `list_integrations` | The sources PodData reads, by family: each with what we read from it, by which route, how often, whether it's live or late or by request, when it last ran, how much we hold and since when, and what its terms say with the page and the day we read them. | Reads | `GET /integrations`, `GET /integrations/{id}` |
| `source_items` | What PodData holds from one source, item by item, newest first: each episode, filing, release, deck, event, record, signal or post, with its title (at most 80 words), its day, its company, show or person, its link on the original's own site, and the id that reads it whole with another tool. | Reads | `GET /integrations/{id}/items` |
| `help` | PodData's own guide to PodData: what trackers, digests, Search, lists, reports, the chat, keys, the MCP server, plans and the rest do, and where in the workspace each is done. | Reads | `GET /help`, `GET /help/{topic}` |
| `call_times` | The coming week's open times for a 30-minute call with PodData's team, a few a day, in the person's time zone, each with the call\_at to book it with ask\_poddata\_team. | Reads | `GET /help/times` |
| `ask_poddata_team` | Sends a question to PodData's team, who answer by email to the asker's address; with call\_at, an open time from call\_times, it also books a 30-minute call then, and Cal.com sends the invite. | Changes something | |
| `list_posts` | What listed people and companies posted in public on their own confirmed accounts (X, Bluesky, Mastodon), newest first, each with who posted it, the network, the day and a link. | Reads | `GET /posts`, `GET /posts/{id}` |
| `list_report_runs` | The runs PodData's own agent has made on one of the team's diligence reports, newest first: a first draft or a change and its message, where each stands, what it cost against its cap, and each agent's part. | Reads | `GET /diligence-reports/{id}/runs` |
| `get_report_run` | One run of PodData's own agent on a report: where it stands, what it cost, and each agent's turns, with the model, the cost and the tools it called in each. | Reads | `GET /diligence-runs/{id}` |
| `start_report_run` | Start PodData's own agent on one of the team's diligence reports: with no message, it writes each page still empty and then the overview, which is every page on a new report and, after a run that ended capped, the pages it left; with a message, the change it asks for, leaving the rest. | Changes something | |

Each operation in the API reference names its tool too, so you can go from a route to its tool and back.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.