"Enhanced MCP-Reasoner with Monte Carlo Tree Search and Beam Search Integration" matching MCP tools:

• retirement_monte_carlo_calculatorA

  calculatorsB

  Retirement Monte Carlo Calculator — Estimate your retirement plan's success probability with 5,000 simulated Monte Carlo market paths, plus a fan chart of percentile balances at every age.

  ConnectorNo auth

• run_monte_carloA

  Synthex Forge — x402 Paid Agent ServicesC

  Return ONE signed, dated Monte Carlo distribution. use it rather than `assess_market_regime` for monte carlo, not market regime; an object with 32 fields including as_of, canonicalization. $0.15 USDC.

  ConnectorNo auth

• list_saved_searchesA

  Moxlade — Upwork buyer intelligenceA

  Every search saved against this token, with whether it is currently matching. Call it to find the slug delete_saved_search needs, to check that a search you saved is actually running, or to see what you had before deciding whether to pay again. Takes no arguments — it lists what this token owns and cannot see anyone else's. Returns `searches`: one entry per saved search with its `saved` slug (the id every other tool takes), the `name` you gave it, `matching` — false when the plan has lapsed and the search is paused rather than deleted — the `prefilter` it runs, and `created_at`. Read-only, and readable on every plan state including expired. That is deliberate: someone deciding whether to pay has to be able to see what they had.

  ConnectorOAuth

• list_saved_searchesA

  Moxlade — Upwork buyer intelligenceA

  Every search saved against this token, with whether it is currently matching. Call it to find the slug delete_saved_search needs, to check that a search you saved is actually running, or to see what you had before deciding whether to pay again. Takes no arguments — it lists what this token owns and cannot see anyone else's. Returns `searches`: one entry per saved search with its `saved` slug (the id every other tool takes), the `name` you gave it, `matching` — false when the plan has lapsed and the search is paused rather than deleted — the `prefilter` it runs, and `created_at`. Read-only, and readable on every plan state including expired. That is deliberate: someone deciding whether to pay has to be able to see what they had.

  ConnectorOAuth

• get_briefingA

  the drainA

  Before security-sensitive tooling or agent operations: find recent threads relevant to a task, with source links, provenance and matching keywords. Keyword overlap on title, tags and excerpt over the last 200 active threads; not semantic search and not the full archive (use search for that). Seeds excluded by default.

  ConnectorNo auth

• riksdag_search_speechesA

  sweden-parliament-mcp-serverA

  Search speeches within a required parliamentary session, optionally by member, party, reply exclusion and heading tokens. Heading matching searches debate headings only, never body text. Search hydrates one text page by default; use get mode with an item speechId to read all remaining text and retain the returned revision for continuation.

  ConnectorNo auth

• query_docs_filesystem_honeydew_documentationA

  Honeydew AI DocumentationA

  Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Honeydew Documentation documentation pages. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

  ConnectorNo auth

• claimhit_rerun_searchA

  ClaimHitA

  Re-run a previous patent search to get fresh results, repeating the inputs that steered it — the company scope, the hint, named products, markets, company sizes and the named standard. Free within 6 months of the original search, 1 search credit after that. Asks first either way: the initial call returns a quote listing the inputs it will repeat and whether it is free, and runs nothing until you call again with confirm: true. To narrow or re-read the results you already have, use claimhit_get_result instead — that is free and does not start a new search.

  ConnectorNo auth

• search_opportunitiesA

  Open Medici Public Fellowship and Grant SearchA

  Search published public fellowships, grants, scholarships, prizes, and funded residencies. For matching, start with separate broad calls for the exact topic, synonyms, broader fields, and type-only grants and fellowships. Do not stack sparse topic, career, and location filters during initial discovery. Array values are OR alternatives within one filter; separate filter fields are combined with AND. Search results are candidates only: call get_opportunity before ranking or citing each one.

  ConnectorNo auth

• query_docs_filesystem_political_commsA

  docsA

  Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Political Comms documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything - `cat /openapi/api-reference/openapi.json | jq '.paths | keys'` — list OpenAPI endpoints OpenAPI specs for this site are mounted at: `/openapi/api-reference/openapi.json`. Use them to answer questions about endpoints, request/response schemas, parameters, and authentication. Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

  ConnectorNo auth

• simulate_portfolio_outcomesA

  Zyberno stock valuationA

  Runs a Monte Carlo simulation over a portfolio balance and returns the distribution of possible end values: percentiles, mean, median, and if you give a goal amount, the share of simulated paths that reached it. Returns are drawn from a normal distribution using the expected return and volatility you supply. Use when the user asks about the range or probability of outcomes rather than a single projected number, for example the odds of reaching a target, or how much volatility widens the spread. Do not use it to value a company (use calculate_dcf_value or get_stock_valuation), and do not use it for retirement withdrawal, tax, or drawdown-sequencing questions, which it does not model. The simulation propagates the assumptions you give it. Normally distributed returns understate real market tail risk, and the output is a property of the inputs, not a prediction about any real portfolio.

  ConnectorNo auth

• grepA

  LayvenA

  Search file contents with a regular expression (RE2 syntax, no backreferences or lookaround). Omit both path and prefix to search every text file in the workspace; pass prefix to search one folder, or path to search a single file. Returns matching lines with 1-based line numbers and surrounding context; line numbers are display-only, never include them in edit's old_str. Matches within single lines only. Use grep to locate exact text; use read to view the surrounding region.

  ConnectorOAuth

• search_sec_event_filingsA

  AlphaCreek SEC Filings MCPA

  Search US SEC filing nodes for a disclosure claim. The search covers 8-K, 6-K, 10-K, 10-Q, and 20-F. Omit document_type to search all of those forms. Pass document_type to search only the forms you name. Do not use this for DEF 14A. How to search: 1. Always pass concept_groups. Every group is required (AND). Within each group's any_of list, one alternative must match (OR). Use 2-5 groups for separate parts of the claim: topic, action, metric, and a causal link when that link is essential. Example: geography + weaker demand + decline, or AI + productivity + margin. Add synonyms. English stemming covers inflections (decline/declined), not synonyms (sales is not revenue). Do not add a group for a detail the filing may leave implicit. 2. query is optional. Each item is an exact adjacent-token phrase. Put alternate full phrasings in the same list. Query plus concept_groups is hybrid search: an exact phrase ranks first, so a generic phrase such as "margin expansion" can outrank the causal sentence you want. Prefer phrases that state the relationship. Do not pass a broad topic word alone. Do not search with query only. Omit query for concept-only search. 3. start_date and end_date are required. tickers limits issuers. Optional document_type limits the search to one or more of 8-K, 6-K, 10-K, 10-Q, and 20-F. Omit it to search all of those forms. How to continue: 4. Each call returns one page. There is no total count. If HAS_MORE is true, call again with cursor set to NEXT_CURSOR and the same concept_groups, dates, query, tickers, ne_tickers, document_type, and match_scope. A cursor from different inputs is rejected. An empty page with HAS_MORE true is not the end: call again with that cursor. Stop when HAS_MORE is false or when later pages are not relevant. 5. If a page is mostly co-occurrence (the terms appear, but not the claim), start a new search with no cursor. Add a concept group or exact causal phrases. If a page is too sparse, start a new search with no cursor and drop the least essential group. Do not change filters while passing a cursor. 6. ne_tickers excludes issuers you have already reviewed. It does not load the next page. Results are candidates. Call read_node_content with document_id and node_id. node_id may start with sp-. Verify negation, causality, comparisons across periods, and numeric thresholds in the source text. Cite CITATION_MARKDOWN.

  ConnectorNo auth

• search_jobsA

  Jobily JobsA

  Search job listings on Jobily.gr, the Greek job board. Filter by free-text term, role/location/sector slugs, company, employment type, workplace type and work time. Use lookup_roles_and_companies, lookup_locations and list_sectors to discover valid slugs — unrecognized slug filters are reported in matchedCriteria.unrecognizedTerms and ignored by the search. Returns up to 20 jobs per page with the total count, facet counts (first page) and recovery suggestions when nothing matches. Search results do not include job descriptions — call get_job with a result's guid for the full posting. When total is large (100+) and few filters are active, recommend the user narrow their search using the facets or lookup tools rather than paginating through hundreds of results.

  ConnectorNo auth

• query_docs_filesystem_site_gpt_docsA

  SiteGPT DocsA

  Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the SiteGPT Docs documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything - `cat '/openapi/https://sitegpt.ai/api/v2/openapi.json' | jq '.paths | keys'` — list OpenAPI endpoints OpenAPI specs for this site are mounted at: `'/openapi/https://sitegpt.ai/api/v2/openapi.json'`, `'/openapi/https://sitegpt.ai/openapi.yaml'`. Use them to answer questions about endpoints, request/response schemas, parameters, and authentication. Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

  ConnectorNo auth

• simulate_priceA

  The Undesirables TCG OracleA

  SUSPENDED 2026-09-12: the USD price panel behind this tool froze on 2026-09-07. The oracle answers 200 {"status": "suspended"} with the reason, resume condition and live alternatives, and does NOT charge. The tool is kept so it resumes automatically the day a live USD series exists. Prefer: loan_terms_preview (graded slabs, live), sports_board (live), or the free Japanese two-sided page /jp/card/{game}/{set}/{card}. SIMULATE a card's price path over a horizon YOU choose (days=30..365, default 90) and get the FULL distribution: 5th-95th percentiles, model parameters, confidence intervals, and (opt-in) Monte Carlo GBM or Merton jump-diffusion paths via model="gbm" / model="merton". Default model is the conformal-calibrated risk forecast. Requires current_price. PAID: $0.015 USDC per call. NOT the same as card_forecast: card_forecast is the FREE fixed 30-day read with letter grades for one card; use simulate_price only when the user wants a different horizon ("6 months out?"), the full percentile curve, or a Monte Carlo model.

  ConnectorNo auth

• query_docs_filesystem_dark_funnelsA

  DarkFunnels documentationA

  Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the DarkFunnels documentation pages. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

  ConnectorNo auth

• reddit_get_postA

  Veezee: LinkedIn people & company data for agentsA

  Fetch full post content for up to 100 Reddit posts by their t3_ ids in one call -- the follow-up loop after reddit_search or reddit_get_subreddit_posts. Costs 4 credits for up to 10 ids, +1 credit per further 10 ids; every post comes back with its full body text. With exactly ONE id you may set detail to 'full' for +4 credits to also get the discussion tree (comments flattened in tree order with depth, about 200 per page, with a comments_cursor to continue), or pass comment_id (a t1_ id, also +4 credits) to fetch one specific comment in its post context. Post ids come from the other Reddit tools or from reddit_resolve_url on a post URL. Not for discovering posts; search first, then batch-fetch here.

  ConnectorOAuth

• add_keywordsA

  GetCitedA

  Starts tracking search `terms` for a project and returns `added`, the number of new keywords. Pass terms as an array of plain search phrases (terms: ["seo tool", "rank tracker"]) with the projectId from add_project or list_projects. Terms are lowercased, and one already tracked is skipped, so calling again is safe. Adding measures no rank: call check_rankings for positions. Search volume and difficulty arrive within about ten minutes, looked up in one request pooled with other keywords, costing one keyword_lookups unit per keyword; read them with list_keywords.

  ConnectorNo auth

• query_docs_filesystem_metabindA

  Metabind DocsA

  Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Metabind documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything - `cat /openapi/bindjs/openapi.json | jq '.paths | keys'` — list OpenAPI endpoints OpenAPI specs for this site are mounted at: `/openapi/bindjs/openapi.json`. Use them to answer questions about endpoints, request/response schemas, parameters, and authentication. Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

  ConnectorNo auth

• query_docs_filesystem_metabindA

  Metabind DocsA

  Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Metabind documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path to `head` or `cat` — a page at the URL path `/some/page` lives at `/some/page.mdx`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Paths are specific to this site — never guess them.** Discover real paths with `tree / -L 2`, `ls /`, or the search tool before reading. If a path does not exist, that only means the guess was wrong; it does NOT mean the topic is undocumented — use `rg -il "keyword" /` to find where it is covered. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /some-directory && ls` or `ls /some-directory`). Do NOT assume that `cd` in one call affects the next call. Examples (replace the placeholder paths with real ones from `tree` or search): - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /some-directory/` — show matches with 3 lines of context around each hit - `head -80 /some/page.mdx` — read the top 80 lines of a specific page - `head -80 /page-one.mdx /page-two.mdx /section/page-three.mdx` — read multiple pages in one call - `cat /some/page.mdx` — read a full page when you need everything - `cat /openapi/bindjs/openapi.json | jq '.paths | keys'` — list OpenAPI endpoints OpenAPI specs for this site are mounted at: `/openapi/bindjs/openapi.json`. Use them to answer questions about endpoints, request/response schemas, parameters, and authentication. Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/some/page.mdx` becomes `/some/page`.

  ConnectorNo auth