Presend MCP Server
io.github.presendapp/presend-mcppresendappstreamable-httpv3.1.2Scanned
Before an AI agent installs an npm/PyPI package: typosquat, vulnerability and existence checks.
40 tools: 1 can delete or cancel things, 28 change things and 11 only read.
- Reads private data · 0
- Reads text others write · 25
- Can send data out · 6
- 9 findings
Tools listed without sign-in. Remote posture
With Anlyon
1 tool here can delete or cancel things. Govern it.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Presend MCP Server can do
The Presend MCP Server is published by the GitHub account presendapp. It has 40 tools. It can delete or cancel things (1 tool: url_clean) and create or change data (28 tools: address_risk, base64, color, csv_json and 24 more). 11 tools only read.
25 tools bring in text other people can write (address_risk, ai_crawler_check, cve_lookup and 22 more); 6 tools can send data out of your system (link_metadata, redirect_trace, rpc_check and 3 more).
The fixed checks matched 9 findings (annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 20, 2026, version 3.1.2.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
40 tools
Deletes or cancels 1Claude Code: deny
url_cleanRemoves 60+ known tracking parameters (utm_*, fbclid, gclid and similar) from a URL and returns the clean URL.2 findings
Why it is in this group
- The description's first verb is "remove": a word for removing or ending something.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- The server marks it not destructive (destructiveHint: false), but an annotation never lowers a class.
Findings
- Warn
The server marks "url_clean" read-only (readOnlyHint: true), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "remove": a word for removing or ending something.
- Warn
The server marks "url_clean" not destructive (destructiveHint: false), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "remove": a word for removing or ending something.
Description, as the server wrote it Removes 60+ known tracking parameters (utm_*, fbclid, gclid and similar) from a URL and returns the clean URL. Does not follow redirects; for that, use redirect_trace.
Annotations the server declares
- title: URL tracking-parameter cleaner
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
Changes things 28Claude Code: ask
address_riskScreens a crypto address against every OFAC SDN digital currency address list.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Screens a crypto address against every OFAC SDN digital currency address list. EVM (0x...) and Bitcoin (bc1..., 1..., 3...) addresses are fully covered (sanctioned true or false, with the matching lists). Addresses of other chains are flagged when listed; Cosmos SDK bech32 addresses return sanctioned: null when not listed, as OFAC publishes none. A sanctions signal only, not a full risk score.
Annotations the server declares
- title: Crypto address sanctions check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
base64Encodes text to Base64 or decodes a Base64 string back to text (action = encode or decode).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Encodes text to Base64 or decodes a Base64 string back to text (action = encode or decode).
Annotations the server declares
- title: Base64 encode / decode
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
colorConverts a color between hex, RGB and HSL.1 finding
Why it is in this group
- The description's first verb is "convert": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "color" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "convert": a word for a change.
Description, as the server wrote it Converts a color between hex, RGB and HSL. Provide exactly one of hex, rgb or hsl.
Annotations the server declares
- title: Color converter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
csv_jsonConverts CSV text to JSON or JSON to CSV (direction: csv-to-json or json-to-csv), for data passed inline.1 finding
Why it is in this group
- The description's first verb is "convert": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "csv_json" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "convert": a word for a change.
Description, as the server wrote it Converts CSV text to JSON or JSON to CSV (direction: csv-to-json or json-to-csv), for data passed inline. CSV must be comma-separated, with a header row and at least one data row; double-quoted fields may contain commas. Semicolon- or tab-separated input is not detected and comes back as a single column. JSON input must be an array of objects: the union of their keys becomes the CSV header and missing values are left empty. Returns result (the converted text), rows and cols. Max 500,000 characters.
Annotations the server declares
- title: CSV / JSON converter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
email_disposableChecks only whether an email address uses a known disposable/temporary email domain.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Checks only whether an email address uses a known disposable/temporary email domain. For syntax, MX, disposable and role-account checks in one call, use email_verify.
Annotations the server declares
- title: Disposable email check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
email_securityAudits a domain's email anti-spoofing setup: SPF strength, DMARC policy and a best-effort DKIM lookup on common selectors.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Audits a domain's email anti-spoofing setup: SPF strength, DMARC policy and a best-effort DKIM lookup on common selectors. A missing DKIM match does not prove DKIM is absent. Checks a domain, not a single address.
Annotations the server declares
- title: Email domain security check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
email_verifyMost complete email check in one call: syntax, MX record, disposable-domain detection and role/generic account detection (e.g.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Most complete email check in one call: syntax, MX record, disposable-domain detection and role/generic account detection (e.g. info@, admin@). Prefer it over email_validate and email_disposable unless you need a single signal. Does not probe the mailbox. valid is null (not false) when the MX lookup could not be completed; retry later instead of treating the address as invalid.
Annotations the server declares
- title: Email verification
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
ip_reputationChecks an IPv4 or IPv6 address against a curated list of netblocks known to be hijacked or run by spam/cyber-crime operations (IPv4-mapped IPv6 uses the IPv4 list).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Checks an IPv4 or IPv6 address against a curated list of netblocks known to be hijacked or run by spam/cyber-crime operations (IPv4-mapped IPv6 uses the IPv4 list). A narrow list-based signal: a clean result is not a safety guarantee. Includes list date and attribution.
Annotations the server declares
- title: IP reputation check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
jwt_decodeDecodes a JWT's header and payload WITHOUT verifying its signature, so its claims must not be trusted on this basis alone.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Decodes a JWT's header and payload WITHOUT verifying its signature, so its claims must not be trusted on this basis alone. To check authenticity, use jwt_verify.
Annotations the server declares
- title: JWT decode
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
jwt_verifyCryptographically verifies a JWT signature (HS256/384/512, RS/PS256/384/512, ES256/384/512) and checks exp/nbf claims.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Cryptographically verifies a JWT signature (HS256/384/512, RS/PS256/384/512, ES256/384/512) and checks exp/nbf claims. Provide a secret for HS*, or a JWK or JWKS URL for RS/PS/ES. Use instead of jwt_decode whenever authenticity matters.
Annotations the server declares
- title: JWT signature verification
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
link_metadataFetches a web page and extracts its title, description, canonical URL, Open Graph and Twitter Card tags and favicon (the data behind link previews).Reads text others writeCan send data out1 finding
Why it is in this group
- The name "link_metadata" says "link": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Findings
- Warn
The server marks "link_metadata" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "link_metadata" says "link": a word for a change.
Description, as the server wrote it Fetches a web page and extracts its title, description, canonical URL, Open Graph and Twitter Card tags and favicon (the data behind link previews). Follows redirects and returns final_url; favicon_source says whether the icon is declared by the page or only the /favicon.ico guess. To see each redirect hop, use redirect_trace.
Annotations the server declares
- title: Link preview metadata
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
maintainer_change_checkPublisher-change analysis is npm only.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Publisher-change analysis is npm only. Also reports whether the package exists (found) and its age (first_published, package_age_days, new_package if first published less than 30 days ago), for npm and for PyPI; on PyPI only existence and age are available. Flags a previously unseen human publisher taking over a package after 180+ days of inactivity, within the last 365 days (the event-stream attack pattern). npm trusted publishing (verified OIDC identity, not just a bot-like account name), pre-release, and handovers to a publisher who already maintains another widely used package (100k+ weekly downloads) are reported but not flagged. Does not detect hijacked existing accounts; a heuristic for review, not proof.
Annotations the server declares
- title: Package maintainer change check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
passwordGenerates a random password, with options for length, symbols, uppercase, numbers and excluding ambiguous characters.1 finding
Why it is in this group
- The description's first verb is "generate": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "password" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "generate": a word for a change.
Description, as the server wrote it Generates a random password, with options for length, symbols, uppercase, numbers and excluding ambiguous characters. To evaluate an existing password, use password_check.
Annotations the server declares
- title: Password generator
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
password_breachChecks whether a password appears in known data breaches (Have I Been Pwned) and how many times, using k-anonymity towards HIBP.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Checks whether a password appears in known data breaches (Have I Been Pwned) and how many times, using k-anonymity towards HIBP. Breach check only; password_check adds strength scoring and sends the password in a POST body.
Annotations the server declares
- title: Password breach check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
password_checkScores a password's strength (length, character variety, entropy as an upper bound, repeated patterns; passphrases are estimated per word) and, with check_breach=true, also looks i…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Scores a password's strength (length, character variety, entropy as an upper bound, repeated patterns; passphrases are estimated per word) and, with check_breach=true, also looks it up in Have I Been Pwned breach data via k-anonymity (only a hash prefix is sent); a password found in breaches is always rated very_weak. Use it to evaluate a password someone is choosing; use password_breach when you only need the breach count, and password to generate a new one. The password travels in a POST body, never in a URL.
Annotations the server declares
- title: Password strength check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
redirect_traceFollows a URL's full redirect chain (up to 15 hops) and returns every hop with its status code, plus whether the chain crossed domains.Reads text others writeCan send data outNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Follows a URL's full redirect chain (up to 15 hops) and returns every hop with its status code, plus whether the chain crossed domains. Use it to see where a short or tracking link really leads; check the final URL with url_reputation.
Annotations the server declares
- title: Redirect trace
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
repo_health_checkMaintenance signals for a GitHub repository given as owner/name: stars, forks, open issues, license, archived and fork flags, creation date and age, days since last push, topics.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Maintenance signals for a GitHub repository given as owner/name: stars, forks, open issues, license, archived and fork flags, creation date and age, days since last push, topics. Use it to judge whether a dependency looks maintained or abandoned. For an npm or PyPI package whose repository you do not know, supply_chain_check resolves it from registry metadata and includes these signals. GitHub only; missing or private repositories return found: false.
Annotations the server declares
- title: Repository health check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
security_headersAudits the HTTP security headers of one URL (CSP, HSTS, X-Frame-Options, Permissions-Policy, cross-origin policies and others) and returns per-header findings with fix advice, a sc…Reads text others writeCan send data outNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Audits the HTTP security headers of one URL (CSP, HSTS, X-Frame-Options, Permissions-Policy, cross-origin policies and others) and returns per-header findings with fix advice, a score and a letter grade. Use it when you need header hardening advice; security_scan runs this audit together with URL reputation and subdomain discovery in one call.
Annotations the server declares
- title: HTTP security headers check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
security_scanCombined website check in one call: security headers, URL reputation and passive subdomain discovery, run in parallel, with an overall score and verdict.Reads text others writeCan send data outNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Combined website check in one call: security headers, URL reputation and passive subdomain discovery, run in parallel, with an overall score and verdict. Use the individual tools when you need a single signal.
Annotations the server declares
- title: Website security scan
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
subdomainsPassive subdomain discovery from Certificate Transparency logs (crt.sh): finds hostnames that appeared in public TLS certificates, not every DNS record.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Passive subdomain discovery from Certificate Transparency logs (crt.sh): finds hostnames that appeared in public TLS certificates, not every DNS record. crt.sh is occasionally slow or unavailable.
Annotations the server declares
- title: Subdomain discovery
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
supply_chain_checkCall this before installing or adding a package (npm install, pip install, a new entry in a manifest), especially one whose name you recalled or that a model suggested.Reads text others write1 finding
Why it is in this group
- The description's first verb is "call": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The server marks "supply_chain_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The description's first verb is "call": a word for a change.
Description, as the server wrote it Call this before installing or adding a package (npm install, pip install, a new entry in a manifest), especially one whose name you recalled or that a model suggested. One-call risk check: combines vulnerability_check (OSV.dev), typosquat_check, maintainer_change_check (npm only) and repo_health_check (when the GitHub repo can be resolved) into one overall verdict. A package that does not exist on npm or PyPI gets overall_risk 'package_not_found': the name may be invented and should not be installed. A package first published less than 30 days ago gets the 'new_package' flag and overall_risk 'review_recommended': new packages are where invented and look-alike names get registered, so the name is worth confirming against the project's own documentation before installing (on PyPI the age is that of the oldest release still published; being new does not make a package malicious). Use the individual tools to investigate one signal. Vulnerabilities are checked for the given version, or the latest published one (version_checked, version_source). If a check could not run (rate limit, upstream error), it is listed in unavailable_checks and overall_risk is 'incomplete', never 'no_signals_found'. Also flags a version marked deprecated on npm or yanked on PyPI.
Annotations the server declares
- title: Package supply-chain check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
text_similarityNear-duplicate detection with a 64-bit SimHash over word shingles: send 1 text to get its hash, or 2 texts to compare them.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Near-duplicate detection with a 64-bit SimHash over word shingles: send 1 text to get its hash, or 2 texts to compare them. Detects paraphrased or lightly edited copies; unrelated texts score around 50%, not 0%.
Annotations the server declares
- title: Text similarity
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
tx_decodeDecodes a raw signed Cosmos SDK transaction (base64 TxRaw bytes, as found in a CometBFT block's data.txs) into JSON: messages, fee, gas, signers and signatures.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Decodes a raw signed Cosmos SDK transaction (base64 TxRaw bytes, as found in a CometBFT block's data.txs) into JSON: messages, fee, gas, signers and signatures. Bank, staking, gov and authz messages are fully decoded; other types are returned as type URL plus raw hex.
Annotations the server declares
- title: Cosmos transaction decoder
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
typosquat_checkCall before installing a package whose name you typed or recalled.1 finding
Why it is in this group
- The description's first verb is "call": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "typosquat_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.
Description, as the server wrote it Call before installing a package whose name you typed or recalled. Checks whether an npm or PyPI package name is a near-miss of a well-known package (typosquatting), with an edit-distance threshold scaled to name length; names of 3 characters or fewer are not fuzzy-matched. Uses a curated list of popular names, so a clean result does not prove a package is safe. It does not check that the package exists: supply_chain_check does.
Annotations the server declares
- title: Package typosquat check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
url_reputationChecks a URL against URLhaus (abuse.ch), a public database of known malware distribution URLs.Reads text others writeCan send data outNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Checks a URL against URLhaus (abuse.ch), a public database of known malware distribution URLs. A clean result only means the URL is not listed, not that it is safe.
Annotations the server declares
- title: URL reputation check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
user_agentParses a User-Agent string into browser and version, operating system and version, device type, and whether it looks like a bot.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Parses a User-Agent string into browser and version, operating system and version, device type, and whether it looks like a bot.
Annotations the server declares
- title: User-Agent parser
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
uuidGenerates 1 to 100 random UUID v4 values.1 finding
Why it is in this group
- The description's first verb is "generate": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "uuid" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "generate": a word for a change.
Description, as the server wrote it Generates 1 to 100 random UUID v4 values.
Annotations the server declares
- title: UUID generator
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
vulnerability_checkChecks a package (optionally a specific version) against OSV.dev for known vulnerabilities: npm, PyPI, Go, crates.io, Maven, RubyGems, Packagist and NuGet.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Checks a package (optionally a specific version) against OSV.dev for known vulnerabilities: npm, PyPI, Go, crates.io, Maven, RubyGems, Packagist and NuGet. For npm and PyPI, a name that does not exist returns found: false and vulnerable: null, never a clean result. Use cve_lookup when you already have a CVE/GHSA ID, or supply_chain_check for a combined verdict.
Annotations the server declares
- title: Package vulnerability check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 11Show 11 read-only tools
ai_crawler_checkFetches a domain's robots.txt and reports which known AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot and others) are allowed or blocked, including wildcard r…Reads text others writeNo findings
Why it is in this group
- The description's first verb is "fetch": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Fetches a domain's robots.txt and reports which known AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot and others) are allowed or blocked, including wildcard rules. Reflects robots.txt only, not server-side blocking.
Annotations the server declares
- title: AI crawler check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
cve_lookupLooks up a vulnerability by identifier (CVE, GHSA or other OSV ID) on OSV.dev: summary, CVSS severity, affected packages and versions, references.Reads text others writeNo findings
Why it is in this group
- The name "cve_lookup" says "lookup": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Looks up a vulnerability by identifier (CVE, GHSA or other OSV ID) on OSV.dev: summary, CVSS severity, affected packages and versions, references. Use when you already have an ID; use vulnerability_check when you have a package name instead.
Annotations the server declares
- title: CVE lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
dns_lookupReturns DNS records for a domain via Cloudflare DNS-over-HTTPS: A, AAAA, CNAME, MX, TXT and NS in one call, or a single record type with 'type'.Reads text others writeNo findings
Why it is in this group
- The name "dns_lookup" says "lookup": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "return" and "http": it reads web pages, which anyone can write.
Description, as the server wrote it Returns DNS records for a domain via Cloudflare DNS-over-HTTPS: A, AAAA, CNAME, MX, TXT and NS in one call, or a single record type with 'type'. For registration data use whois_lookup; for SPF/DMARC/DKIM analysis use email_security.
Annotations the server declares
- title: DNS lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
email_validateLightweight email check: syntax plus confirmation that the domain has an MX record.Reads text others writeNo findings
Why it is in this group
- The name "email_validate" says "validate": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Lightweight email check: syntax plus confirmation that the domain has an MX record. Does not detect disposable or role addresses; use email_verify for the combined check.
Annotations the server declares
- title: Email address validation
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
faviconReturns the favicon URL a website declares: fetches the homepage and takes the first <link rel='icon'> (or 'shortcut icon') href, resolved to an absolute URL, which may be a data: …Reads text others writeNo findings
Why it is in this group
- The description's first verb is "return": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "return" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it Returns the favicon URL a website declares: fetches the homepage and takes the first <link rel='icon'> (or 'shortcut icon') href, resolved to an absolute URL, which may be a data: URI when the page inlines its icon (source: declared). If no icon is declared, or the homepage cannot be fetched, returns the conventional https://<domain>/favicon.ico with source: default and a note, without checking that it exists. Use it to display a site icon; it does not download or validate the image.
Annotations the server declares
- title: Favicon lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
iban_validateValidates an IBAN offline: ISO 7064 mod-97 checksum and country-specific length.No findings
Why it is in this group
- The name "iban_validate" says "validate": a word for reading.
Description, as the server wrote it Validates an IBAN offline: ISO 7064 mod-97 checksum and country-specific length. Confirms the number is well-formed, not that the account exists.
Annotations the server declares
- title: IBAN validation
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
phone_verifyValidates and formats a phone number: validity, country, line type, E.164, international and national formats.No findings
Why it is in this group
- The description's first verb is "validate": a word for reading.
Description, as the server wrote it Validates and formats a phone number: validity, country, line type, E.164, international and national formats. Numbers without a leading + require 'country', since the end user's country cannot be inferred over MCP.
Annotations the server declares
- title: Phone number validation
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
rpc_checkRead-only audit of a public CometBFT (Cosmos SDK) RPC endpoint: node status, health, peers, and whether unsafe admin methods (dial_seeds, dial_peers, unsafe_flush_mempool) are publ…Reads text others writeCan send data outNo findings
Why it is in this group
- The description's first verb is "read": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Claude Code: ask, because it is tagged outbound.
Description, as the server wrote it Read-only audit of a public CometBFT (Cosmos SDK) RPC endpoint: node status, health, peers, and whether unsafe admin methods (dial_seeds, dial_peers, unsafe_flush_mempool) are publicly exposed. Never calls an unsafe method; exposure is inferred from the node's route listing.
Annotations the server declares
- title: Cosmos RPC endpoint check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
timestampReturns the current time, or converts between a Unix timestamp (seconds) and an ISO date.No findings
Why it is in this group
- The description's first verb is "return": a word for reading.
Description, as the server wrote it Returns the current time, or converts between a Unix timestamp (seconds) and an ISO date. Provide unix or date, or neither for the current time.
Annotations the server declares
- title: Timestamp converter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
vat_validateChecks an EU VAT number in real time against the European Commission's VIES service and, when valid, returns the registered company name and address.Reads text others writeNo findings
Why it is in this group
- The name "vat_validate" says "validate": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Checks an EU VAT number in real time against the European Commission's VIES service and, when valid, returns the registered company name and address. VIES is occasionally unavailable for some member states.
Annotations the server declares
- title: EU VAT number validation
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
whois_lookupDomain registration data via RDAP (the modern WHOIS): registrar, creation and expiration dates, domain age in days, nameservers.Reads text others writeNo findings
Why it is in this group
- The name "whois_lookup" says "lookup": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Domain registration data via RDAP (the modern WHOIS): registrar, creation and expiration dates, domain age in days, nameservers. For DNS records, use dns_lookup.
Annotations the server declares
- title: WHOIS lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write25
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out6
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 19:09 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 29
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 29 tools that change things (1 destructive, 28 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://presend.pages.dev/mcp: tools/list returned 40 tools to an anonymous client, including url_clean (destructive), address_risk (write), base64 (write), color (write), csv_json (write) and 24 more
Findings
Matches of fixed checks, with the text they matched.
- Warncolor
The server marks "color" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "convert": a word for a change.
- Warncsv_json
The server marks "csv_json" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "convert": a word for a change.
- Warnlink_metadata
The server marks "link_metadata" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "link_metadata" says "link": a word for a change.
- Warnpassword
The server marks "password" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "generate": a word for a change.
The server marks "supply_chain_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The description's first verb is "call": a word for a change.
- Warntyposquat_check
The server marks "typosquat_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.
Show 3 more findingsShow fewer
- Warnurl_clean
The server marks "url_clean" read-only (readOnlyHint: true), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "remove": a word for removing or ending something.
- Warnurl_clean
The server marks "url_clean" not destructive (destructiveHint: false), and its own name, description or input give the class "destructive".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "remove": a word for removing or ending something.
- Warnuuid
The server marks "uuid" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "generate": a word for a change.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 3.1.2. Values in angle brackets are yours to fill in.
claude mcp add --transport http presend https://presend.pages.dev/mcpclaude mcp add --transport http presend https://presend.pages.dev/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__presend__ai_crawler_check",
"mcp__presend__cve_lookup",
"mcp__presend__dns_lookup",
"mcp__presend__email_validate",
"mcp__presend__favicon",
"mcp__presend__iban_validate",
"mcp__presend__phone_verify",
"mcp__presend__timestamp",
"mcp__presend__vat_validate",
"mcp__presend__whois_lookup"
],
"ask": [
"mcp__presend__address_risk",
"mcp__presend__base64",
"mcp__presend__color",
"mcp__presend__csv_json",
"mcp__presend__email_disposable",
"mcp__presend__email_security",
"mcp__presend__email_verify",
"mcp__presend__ip_reputation",
"mcp__presend__jwt_decode",
"mcp__presend__jwt_verify",
"mcp__presend__link_metadata",
"mcp__presend__maintainer_change_check",
"mcp__presend__password",
"mcp__presend__password_breach",
"mcp__presend__password_check",
"mcp__presend__redirect_trace",
"mcp__presend__repo_health_check",
"mcp__presend__rpc_check",
"mcp__presend__security_headers",
"mcp__presend__security_scan",
"mcp__presend__subdomains",
"mcp__presend__supply_chain_check",
"mcp__presend__text_similarity",
"mcp__presend__tx_decode",
"mcp__presend__typosquat_check",
"mcp__presend__url_reputation",
"mcp__presend__user_agent",
"mcp__presend__uuid",
"mcp__presend__vulnerability_check"
],
"deny": [
"mcp__presend__url_clean"
]
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as presend. Review it before you use it.
Capability tags behind 25 rules
- mcp__presend__address_risk untrusted-input
- mcp__presend__ai_crawler_check untrusted-input
- mcp__presend__cve_lookup untrusted-input
- mcp__presend__dns_lookup untrusted-input
- mcp__presend__email_security untrusted-input
- mcp__presend__email_validate untrusted-input
- mcp__presend__email_verify untrusted-input
- mcp__presend__favicon untrusted-input
- mcp__presend__ip_reputation untrusted-input
- mcp__presend__jwt_verify untrusted-input
- mcp__presend__link_metadata untrusted-input, outbound
- mcp__presend__maintainer_change_check untrusted-input
- mcp__presend__password_breach untrusted-input
- mcp__presend__password_check untrusted-input
- mcp__presend__redirect_trace untrusted-input, outbound
- mcp__presend__repo_health_check untrusted-input
- mcp__presend__rpc_check untrusted-input, outbound
- mcp__presend__security_headers untrusted-input, outbound
- mcp__presend__security_scan untrusted-input, outbound
- mcp__presend__subdomains untrusted-input
- mcp__presend__supply_chain_check untrusted-input
- mcp__presend__url_reputation untrusted-input, outbound
- mcp__presend__vat_validate untrusted-input
- mcp__presend__vulnerability_check untrusted-input
- mcp__presend__whois_lookup untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/io.github.presendapp/presend-mcp/verify?hash=39f03280a55c6ff576c1a50668e44f6f81a75ee509a8d966962ba07235d82242" | jq -e '.changed == false'Pinned hash 39f03280a55c, from Nov 20, 2026, 08:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.presendapp/presend-mcp/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.presendapp/presend-mcp)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:45 UTC.
Presend MCP Server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Presend MCP Server safe?
- No scan can promise that. What its definitions show: 40 tools, 29 of which can change, send or delete something, and 9 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the Presend MCP Server delete data?
- Yes. 1 tool can delete, cancel or revoke something: url_clean.
Can the Presend MCP Server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Presend MCP Server have?
- 40 tools: 1 destructive, 28 write, 11 read. url_clean, address_risk, base64, color, csv_json, email_disposable, email_security, email_verify and 32 more.
Does the Presend MCP Server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Presend MCP Server send data to outside services?
- 6 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): link_metadata, redirect_trace, rpc_check, security_headers, security_scan and url_reputation.
Has the Presend MCP Server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 20, 2026.