Skip to content
Conduit

skill

plivo-skill

Claude skill for Plivo REST API: SMS, voice, WhatsApp, OTP, numbers, subaccounts, with XML and webhook payload references.

priyanshu-plivo/plivo-skillcommunitylistedunreviewed

Set up

Skills are installed by placing their folder where your agent loads skills.

The download comes from github.com/priyanshu-plivo/plivo-skill itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 26, 2026

plivo-skill

A Claude skill for using Plivo's REST API from any agent — SMS, voice, WhatsApp, Verify, numbers, accounts.

What it is

A directory of markdown files that teaches Claude to call Plivo APIs directly. Drop into ~/.claude/skills/plivo/ (or anywhere Claude reads skills) and Claude will use it whenever the user asks about Plivo.

No code, no MCP server, no separate process. Just markdown.

Install

Recommended (via skills.sh CLI):

npx skills add priyanshu-plivo/plivo-skill

Manual install (global, all projects):

git clone https://github.com/priyanshu-plivo/plivo-skill ~/.claude/skills/plivo

Project-local:

git clone https://github.com/priyanshu-plivo/plivo-skill .claude/skills/plivo

Set env vars in your shell:

export PLIVO_AUTH_ID=MAxxxxxxxxxxxxxxxxxx
export PLIVO_AUTH_TOKEN=xxxxxxxxxxxxxxxxxxxxxxxx

Get these from https://cx.plivo.com/home.

Layout

plivo-skill/
├── SKILL.md             entry point (read first by Claude)
├── cheatsheet.md        top-80% ops, one page (fast path)
├── auth.md              auth + env vars
├── safety.md            agent-vs-human tiers
├── not-supported.md     what Plivo doesn't expose via API
├── api/                 per-resource detailed references
└── workflows/           multi-step recipes

Optimization

This skill is designed for low context cost:

  • SKILL.md is ~30 lines — small load when skill triggers
  • cheatsheet.md (~100 lines) covers ~80% of tasks in one Read
  • api/*.md are lazy-loaded — only read the one(s) you need
  • Tables over prose, curl over English — max info density
  • Frontmatter triggers auto-invoke only on Plivo-related intents

Typical cost per task: 150–300 lines of context, vs ~2000 lines if loaded monolithically.

Coverage

All Plivo REST API resources documented at https://plivo.com/docs/.

What's NOT covered

Manual / console-only operations (account recharge, number porting, KYC, WhatsApp template creation, WABA onboarding, team management). See not-supported.md for the full list.

CLI follow-up

This skill is markdown-only — it cannot enforce safety (just instructs Claude to ask). For real enforcement of spend caps, confirmation gates, and dry-run mode, build a thin CLI that wraps the same operations and have this skill call the CLI for any mutation. The skill remains the agent-facing manual.

Contributing

See CONTRIBUTING.md. Issues and PRs welcome.

License

MIT — see LICENSE.

↑↓ navigate openesc close