"Creating a voice AI for making outbound calls via SIP telephony" matching MCP tools:

• ping

  Nordic Financial MCP

  Connectivity check that confirms the Nordic MCP server process is responding. Use this at the start of a session to verify the server is reachable before making other calls. Do not use as a proxy for database health — the server can respond while the Qdrant vector database is temporarily unavailable. To confirm data availability, call search_filings directly. Returns: A greeting string: "Hello {name}! Nordic MCP server is running."

  Connector

• how_munimji_helps

  HelloBooks AI Agents

  Explain how HelloBooks and Munimji (the in-app AI assistant) help a specific business — given a free-text description of the user's own operations. Returns a curated capability knowledge base: business-operation areas (sales, purchases, banking, tax, reports, inventory, payroll, multi-entity, setup), and for each AI capability WHO does the work — `autonomous` (Munimji does it on its own, e.g. OCR extraction, running reports), `approval` (Munimji prepares the entry and you one-click approve before it posts to the ledger, e.g. AI categorization, find-and-match, creating invoices/bills by chat), `assist` (co-pilot, e.g. guided onboarding, voice), or `manual` (a software feature you run yourself). Each capability links to the backing software features. Use this when a user describes their business and asks "how can HelloBooks help me?", "what can the AI do for my shop/practice/agency?", or "what can Munimji do on its own vs what do I approve?". Pass their description in `businessDescription`; optionally filter by `area` or `autonomy`. The AI never posts to a ledger without approval. For the full software catalog call list_features; for pricing call list_plans.

  Connector

• voice

  switch

  Your saved voices — one tool for the whole voice library. Users speak plain language and never know ids: resolve every voice by NAME yourself (call action "list" first if unsure) and never ask the user for an id. action="list" returns every saved voice with voice_id, name, kind and ready — kind "reference" is an instant voice match saved from a clip and kind "clone" is a trained voice (both speak through generate_audio: pass the NAME as its voice param); kind "avatar" voices drive talking_avatar_video. action="create" saves a NEW reference voice from a clip: voice_name plus audio_url (e.g. the url upload_media returned) or audio_base64 (+ format) — free, ready instantly. action="rename" renames a saved voice (voice_id takes the id OR the current name, new_name is the new name). action="clone" registers a voice for talking_avatar_video from audio_sample_url + voice_name (charged 2 credits). action="delete" removes a voice by voice_id or name.

  Connector

• minimize_sip_trace

  Sipflow

  [cost: free (pure CPU, no network) | read-only, no persistence] Reduce a raw SIP trace to a compact form suitable for sending to an LLM. Preserves SDP bodies and routing/auth/dialog headers; prunes well-known noise (User-Agent, Server, Allow, Accept-*, Date, P-* informational, etc.). Expected input format: raw SIP messages separated by blank lines, each starting with a request line (`INVITE sip:...@... SIP/2.0`) or status line (`SIP/2.0 200 OK`). PCAP-decoded text from sngrep / ngrep / tcpdump / tshark, syslog with SIP body, sipflow's own export format, or a hand-pasted INVITE/200 dialog all work. Annotation lines like `# [timestamp] sender -> receiver` or ngrep-style `U <ip>:<port> -> <ip>:<port>` between blocks are tolerated. Safe to run on production traces - the input is processed in-memory and is not persisted or sent off-server. Pair with: `detect_sip_stack` to identify the vendor, then `search_sip_docs(vendor=...)` for vendor-grounded analysis; `render_sip_ladder` to visualize the trace as a Mermaid call-flow ladder; `lint_sip_request` / `parse_sip_message` to mechanically validate any single message in the trace.

  Connector

• dns_diagnose_sip_target

  Sipflow

  [cost: external_io (DNS via Cloudflare + Google; TLS handshake + a SIP OPTIONS keepalive to public targets when applicable) | read-only | rate-limited per IP: 10/min, 200/day] Walk DNS the same way a SIP UA does (RFC 3263 §4.1): NAPTR → SRV → A/AAAA. Given a SIP URI ("sip:example.com"), bare hostname ("example.com"), or "host:port" string, return the records that exist and the resolution ladder a UA would try. When the queried target uses TLS (`sips:` URI, `transport=tls/wss`, or any `_sips._tcp` SRV record), the tool also performs a TLS handshake against each resolved sips target and reports the negotiated TLS version + cipher, the leaf certificate's subject / issuer / SANs / validity, the chain length and whether it validates against Node's default trust store, plus two cert-domain checks: RFC 5922 §7.2 strict (cert must cover the original SIP domain) and a lenient SAN match against the SRV target hostname. SIP liveness: DNS resolving and a TLS handshake succeeding do NOT prove the endpoint actually speaks SIP - a load-balanced node can accept TCP/TLS yet black-hole SIP. So the tool ALSO sends a real SIP OPTIONS keepalive to each resolved public IP across the relevant transports (UDP/TCP on 5060, TLS on 5061 / SRV port) and reports per-IP answered / timeout / refused. Any SIP response (even 405/403/404) proves the stack is alive on that IP. When a name resolves to multiple IPs it is treated as a load-balancer fan-out and each IP is probed individually, with a warning about the known failure modes of fronting stateful SIP/RTP with a cloud L4 LB (AWS NLB/ALB etc.): cross-zone-off targets that black-hole, the ~120s UDP idle timeout, and per-5-tuple hashing splitting signaling from media. Egress safety: - Per-IP rate limited. - Hostnames that resolve only to RFC 1918 / loopback / link-local / documentation / multicast space are refused (SSRF guard). - Walk depth capped to prevent runaway NAPTR / CNAME chains. - TLS probes capped at 6 (host, port, ip) tuples per call, 5 s handshake timeout each, public-IP only (we connect to the resolved IP, not the hostname, so the system resolver cannot redirect us into private space). - SIP OPTIONS probes capped at 6 (ip, transport) tuples per call, 3 s timeout each, public-IP only; the request carries no SDP/body and an unroutable Via, and only the response status line is captured. Use to diagnose: - "carrier doesn't answer" / "wrong port" / "TLS instead of UDP" routing puzzles - "DNS looks healthy but calls fail" - per-IP SIP OPTIONS surfaces nodes that resolve and accept the transport but never answer SIP (the decisive step for load-balanced / multi-IP targets) - "carrier rejects our target because no SRV is published" - when A/AAAA resolves but SRV is missing the tool synthesises a copy-pasteable suggested zone-record block pointing at the resolved canonical hostname - "TLS handshake works but cert isn't valid for the SIP domain" - RFC 5922 §7.2 compliance is checked separately from generic chain validation, since the SAN must cover the *original* SIP domain (not the SRV-redirected target) ACL caveat: a SIP OPTIONS timeout can also mean the target authorizes inbound SIP by source IP whitelist on the trunk (Twilio, Telnyx, Bandwidth, …; see https://www.twilio.com/docs/sip-trunking/api/ipaccesscontrollist-resource) and is dropping our probe because our egress IP is not on the ACL. An `answered` result is conclusive (the node speaks SIP); a `timeout` is suggestive, not proof of a dead node - confirm reachability from the SBC itself. Pair with: `troubleshoot_response_code` when 503 / 408 / 480 are involved; `search_sip_docs(vendor=...)` for carrier-specific routing docs.

  Connector

• fetch_sipflow_share

  Sipflow

  [cost: external_io (Mongo + S3 fetch on the Sipflow backend) | read-only, no persistence | rate limit: shared with the public share endpoint] Given a Sipflow share URL (https://sipflow.dev/share/<token>, or any sipflow.dev subdomain that serves /share/<token>), load the shared SIP trace AND any prior AI analysis attached to it in a single round trip. Use this whenever a user pastes a `/share/<token>` URL: the tool fetches the redacted trace text, the AI executive summary / root-cause / remediation steps (if present), and metadata (vendor, filename, source format, pseudonymized flag), so the agent can review the trace alongside the user's own configs without manual download + paste. In addition to the AI output, the response includes rule-based diagnostics: detected issues (severity-tagged SIP/SDP/media problems with RFC references), WebRTC signal checklist scores, multi-leg call correlation (Session-ID grouping), and detected SIP stacks (User-Agent/Server header values). These diagnostics are computed at share-creation time; for older shares without persisted diagnostics, the tool parses the trace on the fly. When the share includes media quality data (from PCAP-sourced captures), the response includes per-call MOS/jitter/loss summaries in the text output and full `mediaQuality` stats in `structuredContent`. If `hasRawCapture` is true, the sharer included their original PCAP for full RTP playback on the web UI - this raw binary is not returned to agents. Privacy: the share endpoint deliberately strips the original `problem` and `architecture` fields the sharer typed in (those may contain customer-internal context). This tool returns the same public projection - only the trace, the AI output, diagnostics, and basic metadata. Traces are pseudonymized by default (phone numbers / IPs / Call-IDs replaced with consistent fakes); the `pseudonymized` field tells you whether the sharer opted to keep raw values. Trace bytes are capped at 200kB (matching the budget the Sipflow AI worker uses). For very large captures the response sets `trace.truncated=true` - pair with `minimize_sip_trace` to compact further before passing to your own LLM, or with `render_sip_ladder` to visualize the call flow. Pair with: `review_sip_config` to compare the shared trace against the user's own kamailio.cfg / pjsip.conf / FreeSWITCH XML; `render_sip_ladder` to draw the shared call flow inline; `minimize_sip_trace` if `trace.truncated` is true; `troubleshoot_response_code` for any failing transactions surfaced in the AI analysis.

  Connector

• createVoice

  Ludo AI Game Assets

  Design a new voice from a character description (such as "deep-voiced warrior" or "cheerful young girl") and have it speak a short line of text, returning a sample of that newly created voice. Both voice_description and text are required, the spoken text is limited to 200 characters or the call returns HTTP 400, and type selects "human" or "non-human" voices. The job result is a single audio result containing a URL; there is no separate polling step. Credits are charged on success. Use this to invent and audition a voice from a description; use createSpeech for text-to-speech that clones a specific voice from an audio sample, and createSpeechPreset for text-to-speech using a named preset voice. Pass an optional request_id to tag the result so you can locate it later via `GET /assets/audio/results`. Requires an API key (user scope). Returns 202 with a job id immediately; poll `getApiJob` (pass `wait: 30`) until status is succeeded, then read its `result` field, which is exactly the response documented for this operation. Each account may have up to 50 generations queued or running at once; beyond that submissions return 429 (PENDING_JOBS_LIMIT) - wait for jobs to finish. Credits: This endpoint consumes 1 credits per call.

  Connector

• generate_audio

  switch

  Generate spoken audio from text: narration, a voiceover, a read-aloud script, or a multi-voice dialogue. Pass text (up to 2048 chars) — the words to be spoken. To speak in one of YOUR saved voices, pass voice with the voice NAME (or id): users speak plain language and never know ids, so resolve the name yourself (the voice tool, action "list", shows every saved voice) and never ask the user for an id. Reference voices, trained clones and preset voices are all routed correctly by kind. To match a voice instantly from a clip instead, pass reference_audio_url (a short clip) or up to 3 reference_audio_urls and address them as @Audio1, @Audio2, @Audio3 in the text for dialogue. Alternatively pass image_url to voice a scene from a picture (cannot combine with reference audio). Optional speech_rate (-50..100), pitch (-12..12), loudness (-50..100). Returns a playable audio_url, duration_seconds, and generation_id (also saved to your library).

  Connector

• get_category_landscape

  StackSwap

  Full map of one GTM category — leaders, runner-ups, and skip/replace candidates. Returns every catalogued tool in the bucket with cost, AI-readiness, swap-registry status, and partner sign-up links. Use when the user wants to see the full landscape for a category (e.g. 'show me all CRMs', 'what outbound tools exist', 'map the analytics category') — strictly more comprehensive than `recommend_partner` (single best pick). Known buckets: crm, outbound, data, marketing-automation, analytics, meetings, support, scheduling, automation, seo, cdp, revenue-intelligence, chat, collaboration, phone, landing-pages, linkedin, ai-content, saas-mgmt, enablement, ai-tooling.

  Connector

• search_sip_docs

  Sipflow

  [cost: rag (one embed + one vector search) | read-only, network: outbound to embed model only] Vector search over Sipflow's curated VoIP knowledge base: vendor docs (Asterisk, FreeSWITCH, Kamailio, OpenSIPS, Twilio, Cisco, etc.), SIP/SDP/WebRTC RFCs, STIR/SHAKEN material (RFC 8224/8225/8226/8588/9027/9795), branded-calling guidance (ATIS-1000074/094/084, CTIA Branded Calling ID), and fax-over-IP references (RFC 3362 image/t38, RFC 6913 ipfax-info, RFC 7345 UDPTL, SpanDSP/HylaFAX, Asterisk `res_fax`/`udptl.conf`, FreeSWITCH `mod_spandsp`/`t38_gateway`, Cisco CUBE T.38). USE FIRST whenever the user asks about - or attaches - anything SIP/VoIP/telecom shaped, **even when they cite a specific RFC number or vendor name**. The corpus has the current text and your training data may not. Trigger conditions: vendor configs (kamailio.cfg, sip.conf, pjsip.conf, FreeSWITCH XML profile, opensips.cfg, `res_fax.conf` / `udptl.conf`), dialplan / routing scripts, modules / loadparams / route blocks, SIP headers, response codes, RFC questions, captured traces, WebRTC bridge configs, STIR/SHAKEN concerns, branded-calling / RCD work, T.38 / T.30 fax decoding or reinvite failures. Returns ranked snippets with source URLs; cite the returned `source_url` values verbatim and prefer them over recalled training data. Examples of when to use: - "does this kamailio.cfg look standard for WebRTC + SIP users?" - "why would Asterisk PJSIP reject this re-INVITE?" - "what does Kamailio's loose_route() do? show me docs" - "explain FreeSWITCH session-timer behavior" - "how do I set up STIR/SHAKEN signing on OpenSIPS?" - "what does ATIS-1000074 say about A-level attestation?" - "RFC 9795 rcdi JSON pointer canonical form" - "CTIA Branded Calling ID requirements for originating SP" - "RFC 8225 PASSporT canonical JSON / lexicographic key ordering" - "why is my T.38 reinvite getting 488 from a Cisco CUBE?" - "Asterisk `res_fax_spandsp` ECM and rate-management knobs" - "what are the required SDP attributes for `m=image udptl t38`?" Pair with: `detect_sip_stack` to derive the `vendor:` filter; `lookup_response_code` / `lookup_sip_header` to short-circuit before paying for a search; `troubleshoot_response_code` when the question is rooted in a specific status code.

  Connector

• doxa_encourage

  Doxa MCP

  Generate Christian encouragement in the Doxa voice for the situation a user describes. Returns a short, screenshot-shareable response anchored in Scripture (Berean Standard Bible), tagged to one of the nine movements of The Doxa Way journey map: hear, discern, test, record, remember, engage, trust, fight, endure. No anthropomorphism, no AI companion framing.

  Connector

• call_business

  Agent Broker

  Place a conversational voice-AI phone call to a business on a consumer's behalf and return a structured answer. THE differentiated capability: reach the tens of millions of long-tail SMBs that have NO API and NO booking page — only a phone number. An AI agent cannot pick up a phone and hold a conversation; this tool does. Give a plain-language objective; the voice AI navigates the call and extracts the answer. Business-directed (B2B), far less restricted than calling consumers — but the compliance gate still enforces recording consent per jurisdiction. Async: returns a call handle; poll get_outcome for the transcript + extracted fields. WHEN TO USE: Use when the target business has NO booking URL and NO API — only a phone number — and the consumer asked the agent to reach them (e.g. 'call this plumber and ask if they can come Tuesday', 'ask the salon if they take walk-ins this afternoon'). Also use to confirm details a booking page doesn't expose (real-time availability, custom quotes). WHEN NOT TO USE: Do NOT use when the business has a booking URL — use import_booking_url + schedule_appointment (cheaper, faster, deterministic). Do NOT use for calls to consumers/individuals (this tool is for reaching businesses). Do NOT use for marketing or telemarketing — the compliance gate and the B2B-only framing reject that. COST: $0.2 per_call LATENCY: ~45000ms EXECUTION: async_by_default (use get_outcome to retrieve result)

  Connector

• meta_capabilities

  Colour Memory

  Return a live inventory of all active endpoints and MCP tools. Use this first to discover what the API can do before making calls. Returns tool count, endpoint list, MCP-exposed tools, and usage notes. Deterministic -- no LLM cost.

  Connector

• synthesize_voiceover

  TestMyVibes

  Generates a voiceover from text using Hume Octave TTS. Audio uploaded to Spaces, signed URL returned (24h TTL by default). Charged in credits up-front based on script length (use quote_voiceover for a preview). Best for demo-video narration, tutorial audio, and any one-shot batch TTS. NOT a real-time conversational voice (use Hume EVI for that, different product). Voice options: pass voiceId for a specific Hume voice clone, or omit to use the deployment's default narrator (HUME_OCTAVE_VOICE_ID env var).

  Connector

• axiorank_author_detector

  AxioRank: Zero-Trust for AI Agents

  Describe a risk in plain English and AxioRank's AI proposes a custom content detector (regex or keyword) and SAVES it DISABLED for your review. An LLM never arms detection unattended. Outbound content categories only (secret/pii/destructive/injection/egress). Requires the `policies:write` scope and an AI-assessments-enabled plan (Team+).

  Connector

• getThread

  email

  Get a conversation thread by id: metadata plus all inbound and outbound messages interleaved oldest-first. Each message has a direction (inbound/outbound) and id; fetch inbound message bodies via getEmail, or outbound bodies via getSentEmail. Discover thread_id from any email or sent-email record. Compare message_count against messages.length to detect truncation.

  Connector

• security_fetch_cve_watch_status

  DataNexus MCP

  Check all specified CVE watches for new events since your last poll. Returns only watches with new events, making it efficient to run on a schedule. watch_ids: List of watch IDs to check — same IDs used when creating watches with security_fetch_cve_watch. Required. Uses a per-user cursor (last_polled timestamp) stored in Redis. First call returns events from the last 30 days. Subsequent calls return only events newer than the last poll. Sources: Redis (existing watch data written by security_fetch_cve_watch). No external API calls — instant response. If this tool's response does not serve the user's need, call report_feedback with feedback_type="agent_gap", tool_id="security_fetch_cve_watch_status", intended_query="{what the user needed}", gap_description="{what was missing or wrong in the result}".

  Connector

• create_series

  Misar.Blog MCP Server

  Create a new, empty series — a named collection that related articles can be added to. Creating the series does not move any article into it; follow up with add_to_series for each one. Call get_series first to avoid making a second series with the same title, since each call creates a NEW series and nothing deduplicates them. Requires an API key. The series and its URL become publicly reachable, though it shows nothing until articles are added. Returns the series with the slug that add_to_series needs.

  Connector

• generate_audio

  mcp

  Speech and sound: text-to-speech (default; optional voice_id, see list_resources), sound effects (model 'sfx', optional duration), multi-voice dialogue (model 'dialogue': pass the turns in the dialogue parameter, not prompt), transcription (model 'stt': pass audio_url, get text back; optional diarize/language_code), voice change (model 'voice-changer': audio_url + target voice_id) and audio cleanup (model 'voice-isolation': audio_url). Returns the audio URL (or the transcript for 'stt'). For music, use generate_music.

  Connector

• get_api_info

  govbid-mcp-server

  Get metadata about the GovBid Global API including version, data source, license information, and usage guidelines. Call this first to understand the service before making other tool calls.

  Connector