Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add vobiz-ai/Agent-Skills --skill vobiz-voice-callsgit clone --depth 1 https://github.com/vobiz-ai/Agent-SkillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/vobiz-ai/agent-skills/vobiz-voice-calls)<a href="https://agentmods.dev/skills/vobiz-ai/agent-skills/vobiz-voice-calls"><img src="https://agentmods.dev/badge/skills/vobiz-ai/agent-skills/vobiz-voice-calls/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vobiz-ai/agent-skills/vobiz-voice-calls"><img src="https://agentmods.dev/badge/skills/vobiz-ai/agent-skills/vobiz-voice-calls.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00043 | $0.01526 |
| Opus 5 | $0.00022 | $0.00763 |
| Sonnet 5 | $0.00009 | $0.00305 |
| Haiku 4.5 | $0.00004 | $0.00153 |
Grade A, and why
vobiz-voice-calls scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vobiz Voice Calls skill
Use this when the user wants to dial out, control an active call, or look up a live call. For completed-call history, prefer the vobiz-cdr skill.
What you can do
- Make a call -
POST /api/v1/Account/{auth_id}/Call/with{from, to, answer_url, answer_method}. Returns acall_uuid. - List live calls -
GET /api/v1/Account/{auth_id}/Callto see all in-progress/queued calls. - Get one live call -
GET /api/v1/Account/{auth_id}/Call/{call_uuid}. - Hangup a call -
DELETE /api/v1/Account/{auth_id}/Call/{call_uuid}. - Mid-call actions - play audio, speak TTS, send DTMF, start/stop recording, start audio stream. Each is a
POSTto/Call/{call_uuid}/{action}/. - Machine detection - outbound calls support synchronous and async answering-machine detection with configurable sensitivity. See
call/machine-detection.
Required inputs
from: a Vobiz number you own (or a verified caller-ID).to: E.164 format (+91...). Use<to bulk-dial up to 1000 destinations in one request (e.g.14157654321<14153464321<sip:[email protected]).answer_url: HTTPS endpoint that returns VobizXML when the callee answers.answer_method: HTTP verb foranswer_url. The make-call schema treats this as required alongsidefrom/to/answer_url; default toPOST.
Leg and casing semantics (memorize these)
legs/legselects who an in-call action hits:aleg= caller (the A-leg, yourfromside),bleg= callee,both= everyone. Default isalegeverywhere.- Play and Speak use the plural field
legs(aleg/bleg/both). DTMF uses the singular fieldlegand acceptsaleg/bleg/bothper the API schema (some examples show onlyaleg/bleg). request_uuidandcall_uuidare the same identifier returned by make-call; use either.- The base host is
https://api.vobiz.ai; every path is prefixed with/api/v1.
Pitfalls
- Path casing matters:
/Account/.../Call/(uppercaseAccount, uppercaseCall) for call ops. Lowercase paths (/account/.../call/) return 401. - Trailing slash matters and the two list routes differ by it:
GET /Call?status=livelists live calls;GET /Call/?status=queuedlists queued calls.POST /Call/(with slash) makes a call. When in doubt, copy the exact path from the matching doc page. statusis a required query param on every retrieve endpoint (liveorqueued). Omitting it returns the finalized CDR or a 404, not the live record.- Transfer is NOT a REST endpoint in the OpenAPI spec - it is
POST /Call/{call_uuid}/documented incall/transfer-call. Don't confuse it withGET /Call/{call_uuid}/(retrieve queued). The transfer body usesaleg_url/bleg_urland the plurallegsselector; the new URLs must return VobizXML. - Conferences are created on join via XML, not via REST. There is no "create conference" POST - a caller enters a
<Conference>NAME</Conference>room and Vobiz auto-creates it. The REST Conference endpoints only inspect/control/record existing rooms. member_idaccepts a single id, a comma-separated list (10,15,22), orall. Conference names with spaces must be URL-encoded (My%20Conf%20Room); names are case-sensitive.- Play
urls, Speaktext, member Playurlare required; sending them empty fails. Audio must be public.mp3/.wavover HTTP(S). - Stop actions are idempotent: stopping play/speak/recording when nothing is active still returns success (204). Hangup and conference deletes are destructive and not reversible.
- A real call costs money and can fail for 402 (insufficient balance) or be throttled by 429 (CPS/concurrency limit). In sandbox/test work, place at most one short call per scenario.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 62 lines · 43 tokens per session scan A bf1cea38e98c
vobiz-voice-calls is a skill published in the GitHub repository vobiz-ai/Agent-Skills (2 stars, last pushed 23d ago), licensed MIT. It adds 43 tokens to every session and 1,526 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
init-rpm
Project setup and verification. First run creates rpm context for a project. Repeat runs verify that an existing rpm setup matches the latest expected layout and apply safe migrations.
backlog
Manage the rpm backlog (long-term project tasks in docs/rpm/future/tasks.org — distinct from Claude's native TaskCreate list, which is session-scoped). Add, list, review, postpone, or complete entries. TRIGGER on natural-language backlog operations — phrasings like "backlog X", "add X to backlog", "add to backlog"…
zig-docs-mcp
Connect Prime Agent to the local zig-docs-mcp MCP server for always-fresh official Zig documentation of the latest release, std-library symbol docs from released sources, curated high-performance lightweight-software guidance, and safe local Zig toolchain auto-update prompts. Use when writing, reviewing, or debugging…
version
Report the installed rpm plugin version. Use when the user asks for the rpm version, plugin version, installed version, or wants to verify which rpm release is loaded.
zig-docs
Operating rules for working WITH Zig itself: before writing or reviewing Zig code, fetch fresh docs for the latest release via the zig-docs-mcp tools (zdoc singleton), check whether the local toolchain is behind and offer a gated auto-update, and ground all performance guidance in the bundled high-performance…
fpf-reference-mcp
Install, connect, validate, or self-host the FPF Reference MCP server. Use for hosted remote setup, local stdio or HTTP runtime setup, and local bridge/proxy setup for clients that need a localhost MCP URL while using the hosted remote endpoint.