--- name: hashgraph-online/calle source: https://app.decimal.ai/s/hashgraph-online-calle@1/SKILL.md source_sha256: 5369ac075f8a --- # CALL-E Use this skill when the user wants Codex to use CALL-E through the `calle` CLI. This plugin version intentionally calls the CLI instead of configuring Codex to connect directly to the remote MCP server. ## When to use Use this skill for: - verifying CALL-E setup in Codex - checking whether the `calle` CLI is available - recovering from missing or expired CALL-E authentication - listing available CALL-E MCP tools through the CLI - planning a phone call - running a planned call after planning returns complete run credentials - checking a call run status - reporting the final call summary, details, and transcript when a call reaches a terminal status Do not use this skill when the user only wants a call script, roleplay, simulated conversation, or general contact lookup that does not require CALL-E. ## Tool routing When this Codex plugin skill is active, use only the `calle` CLI flow documented below. Do not call ChatGPT App or connector tools, including tool namespaces prefixed with `mcp__codex_apps__`, even if a ChatGPT App has the same visible name, tool names, or MCP service behind it. If a same-name ChatGPT App is available, treat it as a separate integration. This Codex plugin still routes through the local CLI so Codex plugin authentication, attribution, and safety behavior remain isolated from ChatGPT App execution. ## Safety and consent - Real phone calls may contact external people or businesses. - Do not place a real call unless the user clearly intends to do so. - Always plan first. - If the user asked to place a call, run it immediately after planning returns a valid `plan_id` and `confirm_token`. - If the user asked only to verify setup or only to plan, do not run the call. - Do not guess phone numbers, country codes, language, region, `plan_id`, `confirm_token`, or `run_id`. - Do not print, request, or expose access tokens. ## CLI selection All CLI commands run from this Codex plugin must include the CALL-E integration attribution environment: ```bash env CALLE_SOURCE=codex CALLE_INTEGRATION=codex_plugin CALLE_INTEGRATION_VERSION=0.1.11 ``` Use the first command form that works. Prefer the repository-local CLI when the current workspace contains it: ```bash env CALLE_SOURCE=codex CALLE_INTEGRATION=codex_plugin CALLE_INTEGRATION_VERSION=0.1.11 node packages/cli/bin/calle.js ``` If the repository-local CLI is unavailable, use the global command: ```bash env CALLE_SOURCE=codex CALLE_INTEGRATION=codex_plugin CALLE_INTEGRATION_VERSION=0.1.11 calle ``` If neither command works, use the npm package through `npx`: ```bash env CALLE_SOURCE=codex CALLE_INTEGRATION=codex_plugin CALLE_INTEGRATION_VERSION=0.1.11 npx -y @call-e/cli ``` Only tell the user to install the CLI globally if `npx` is unavailable, network access is blocked, or the user explicitly wants a persistent global command. ## Readiness flow Use this flow whenever this Codex plugin is actively invoked for a CALL-E request. Run it before call planning, before tool listing, when setup is uncertain, when auth fails, or when the user asks to verify CALL-E setup: 1. Check CLI availability with `--help`. 2. Run `auth status`. 3. If `auth status` reports `usable: false`, or if this flow is running after any command returned `auth_required`, do not continue to call planning or `mcp tools` yet. Run blocking `auth login` and keep that command running until it exits. If the preceding command returned `auth_required` while `auth status` still reported `usable: true`, add `--force-login` so the CLI does not rely on a locally usable but server-rejected token. Do not use `auth login --start-only --no-browser-open` for the default Codex plugin flow. 4. When `auth login` prints the brokered login URL to command output or stderr, immediately show the first authorization help with that URL. Keep waiting for the same `auth login` command to complete; do not ask the user to reply after browser authorization. 5. If the successful `auth login` JSON included `assistant_hint.message`, show that post-auth success message in the next user-facing reply. If the user already gave a call goal, continue the original workflow after the message; otherwise ask for the phone number and call goal, or offer a test call. 6. After login completes, run `mcp tools`. 7. Confirm that `plan_call`, `run_call`, and `get_call_run` are available. Setup verification must not place a real phone call. Use only help, auth, and tool-listing commands until the user asks for a call workflow. First authorization help template: ```text Hi, I'm CALL-E 👋 I can help you make phone calls, ask for information, and handle phone-related tasks. I'll also keep you updated on the call status, what was discussed, and the key points. Before we officially begin, I'll send you the call goal for confirmation. Before we start, please complete authorization here: ``` Post-authorization success template: ```text Great, authorization is complete ✨ - If you already shared the call goal, I'll continue as planned. - If you haven't, that's okay. I can help you place a test call first, or start a real call directly. You can tell me: - Your phone number: Used only for this service. We will not disclose it to anyone else, including the callee. - What you want me to say: For example, "This is a test call from CALL-E. Wishing you a good day, and asking if there's anything you'd like to share." I'll keep you updated on the phone status, call content, and summary. ``` ## Call flow 1. Use `call plan` first. If the user has not provided enough explicit fields for `call plan`, use `mcp call plan_call --args-json '{"user_input":""}'` so CALL-E can ask for the missing details. 2. Read the returned `plan_id` and `confirm_token`. 3. If the user's request is to place a call, immediately use `call run` with the exact `plan_id` and `confirm_token` returned by planning. 4. Do not ask for a second confirmation between `call plan` and `call run`. 5. Read the returned `run_id` and latest call status. In `call run` output, the latest call state is in `status_result.structuredContent`. In `call status` output, the latest call state is in `result.structuredContent`. 6. If the latest status is not terminal, immediately show a user-visible progress update from the latest activity data before polling again. Use `status_result.structuredContent.activity` after `call run`, or `result.structuredContent.activity` after `call status`. 7. Keep using `call status` with that exact `run_id` until the call reaches a terminal status or the user asks you to stop. Poll every 10 seconds: after each non-terminal response, show the latest activity progress, wait 10 seconds, then fetch `call status` again. Do not stay silent until a terminal status. 8. Use `call status` only with a known `run_id`. Terminal statuses include `COMPLETED`, `FAILED`, `NO_ANSWER`, `DECLINED`, `CANCELED`, `CANCELLED`, `VOICEMAIL`, `BUSY`, and `EXPIRED`. For non-terminal statuses, reply with progress in this shape: ```text Phone call is in progress! Progress: - ``` Use one bullet per `activity` item, preserving the order returned by the CLI. For each item, prefer the event `ts` formatted as `HH:MM:SS` plus `message`. If `ts` is missing, use the message by itself. If there is no activity, use `- Status: ` when a status exists; otherwise use `- Waiting for the next status update.` Do not include the final summary, details, or transcript until a terminal status is returned. The polling cadence is: show progress, wait 10 seconds, run `call status`, show new progress if still non-terminal, then repeat. Stop polling immediately when the user asks you to stop, when a terminal status is returned, or when command execution is interrupted. When the call reaches a terminal status, reply with the final call result, including these sections in this order: ```text [Status] [Call Summary] [Details] Callee Number: Duration: Time: Call id: [Transcript] ``` If the user asked for extra final content, such as key takeaways or next steps, add it after `[Transcript]` under a short heading. Base all final sections only on the JSON returned by `call run` or `call status`; do not invent a transcript. If any command returns `auth_required`, switch to the readiness flow, complete fresh login, and then retry the original operation after login completes. Use `references/commands.md` for exact command examples, supported options, and JSON handling rules.