Agent skill

Language Bridge Call

by CALLE-AI in CALLE-AI/awesome-phone-call-agents

Cross-language relay call skill for CALL-E - places one call in the recipient's language to capture a structured answer, then one call back to the requester in their language, and returns an agreed…

MITAuto-check passed

Install Language Bridge Call

skills CLI
$ npx skills add CALLE-AI/awesome-phone-call-agents --skill language-bridge-call -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install CALLE-AI/awesome-phone-call-agents language-bridge-call --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/CALLE-AI/awesome-phone-call-agents.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/language-bridge-call .claude/skills/language-bridge-call && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
language-bridge-call
GitHub stars
107
Token cost
~2.3k tokens
SKILL.md length
824 words
Files
7 (incl. scripts, references, assets)
Skills in repo
101
Repo updated
First seen
Licence
MIT

At a glance

Cross-language relay call skill for CALL-E - places one call in the recipient's language to capture a structured answer, then one call back to the requester in their language, and returns an agreed…

  • Works in 4 steps: Confirm the user authorized this one… → Confirm both phones are E.164 and came… → Refuse if do_not_call is true or consent… → …
  • SKILL.md covers When To Use, When Not To Use, Required Inputs and Preflight, plus 6 more sections
  • Runs Python scripts from its folder; calls python3

What it does

Language Bridge Call is an agent skill from CALLE-AI/awesome-phone-call-agents. Cross-language relay call skill for CALL-E - places one call in the recipient's language to capture a structured answer, then one call back to the requester in their language, and returns an agreed outcome or a fail-closed handoff.

Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 10 other files, including scripts, reference files and assets (for example `assets/sample-relay-request.json`, `references/examples.md` and `references/safety.md`).

The repository describes itself as: Portable phone-call Agent Skills, apps, examples, adapters, and scheduler recipes for AI agents. The licence is MIT.

Example prompts

  • “/language-bridge-call”

Requirements

  • Python 3

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Confirm the user authorized this one relay (both numbers) for this request.
  2. Confirm both phones are E.164 and came from the requester's own records.
  3. Refuse if do_not_call is true or consent is not true.
  4. Run the dry-run preview and review both goal scripts before any CALL-E plan.

What it can do on your machine

Read from SKILL.md and the folder at commit 38d4118. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 3 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Language Bridge Call loads about 2.3k tokens when it runs, and up to ~3k if it reads all its reference files. Until then it costs about 63 tokens; SKILL.md has 824 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~63
When it runs · the whole SKILL.md, loaded when a task matches
~2.3k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from CALLE-AI/awesome-phone-call-agents at commit 38d4118, republished under its MIT licence (© CALLE-AI). 824 words, ~2,290 tokens.

Download SKILL.mdSave it as .claude/skills/language-bridge-call/SKILL.md (or your agent's skills folder). This skill also uses 6 other files; get the full folder from GitHub.
name
language-bridge-call
description
Cross-language relay call skill for CALL-E - places one call in the recipient's language to capture a structured answer, then one call back to the requester in their language, and returns an agreed outcome or a fail-closed handoff.
license
MIT

Language Bridge Call

Use this skill when a requester (a property manager, workshop owner, coordinator) needs to coordinate a concrete, non-urgent task with someone who does not speak the requester's language well enough for a phone conversation. The skill places at most two disclosed CALL-E calls: first to the recipient in the recipient's language to deliver the request and capture a structured answer, then (only if that answer is readable) to the requester in the requester's language to relay the answer and pin an agreed outcome.

The value is the pair of calls, not a translation document: many recipients answer the phone but not email, forms, or text in a second language. The result is fail-closed structured JSON, not an agreement made on the agent's own judgment.

This skill does not interpret live conversations, certify anything, or commit either party to contracts, payments, or medical, legal, or financial terms.

When To Use

  • Ask for a maintenance access window from a tenant in their language, and report their answer back
  • Confirm a schedule change with a seasonal worker or contractor who prefers another language
  • Relay one bounded request (an object to bring, a document to have ready, a time to arrive) and capture a yes/no or a counter-proposal
  • Produce an evidence-backed, two-sided disposition for a coordinator who would otherwise make phone calls through a friend or a dictionary

When Not To Use

  • Medical, legal, financial, emergency, insurance-claim, or government-interaction interpretation
  • Anything that needs live back-and-forth between both humans on one line (use a human interpreter)
  • Collecting payment details, ID numbers, signatures, or health information
  • Calling a number the user did not provide and authorize in the request
  • Hidden retries, recurring reminders, surveys, marketing, or political outreach
  • Any request where consent from either party is missing or the request is urgent

Required Inputs

  • request_id: stable local identifier
  • topic: short noun phrase for the call (for example "maintenance access")
  • message_to_recipient: what the requester wants conveyed, in plain requester-language sentences
  • requester: display_name, role, first_name, phone (E.164, authorized), language, region
  • recipient: first_name, phone (E.164, authorized), language, region
  • timezone: IANA
  • consent: must be true
  • authorized_reason: why these two specific numbers may be called for this request

Optional: proposed_windows (max 4 ISO-8601 times with offset the requester can actually honor).

Preflight

  1. Confirm the user authorized this one relay (both numbers) for this request.
  2. Confirm both phones are E.164 and came from the requester's own records.
  3. Refuse if do_not_call is true or consent is not true.
  4. Run the dry-run preview and review both goal scripts before any CALL-E plan.

Dry-Run Preview

From the repository root (no CALL-E credentials, no network):

bash
python3 skills/language-bridge-call/scripts/relay_call.py --request skills/language-bridge-call/assets/sample-relay-request.json

Preview validates the request, prints masked phone numbers, both goal scripts, and the exact live command lines. It does not dial.

Fixture mode runs the complete two-call relay state machine against canned CLI envelopes (still no network, no calle install needed):

bash
python3 skills/language-bridge-call/scripts/relay_call.py --request skills/language-bridge-call/assets/sample-relay-request.json --fixture skills/language-bridge-call/scripts/fixtures/relay_happy_path.json
python3 skills/language-bridge-call/scripts/test_relay_call.py

CALL-E Goal Templates

Recipient call, rendered and spoken in recipient.language:

text
You are an AI phone assistant working for {requester_display_name}, a {requester_role}.
Disclose immediately that you are an AI assistant and that this is one coordination call about {topic}.
Speak {recipient_language} throughout. If the person who answers is not {recipient_first_name}, apologize and end the call.

Purpose: relay this request and capture an answer.
Message: {message_to_recipient}
Offer only these windows, one at a time: {proposed_windows}.

Ask {recipient_first_name} to choose a window, propose a different time, or decline.
Do not discuss payments, contracts, deposits, or legal matters. Do not promise anything.
Close politely once you have captured the answer.

Before closing, record the answer as JSON with exactly these fields: understood (boolean),
consent_given (boolean), wrong_person (boolean), confidence (number between 0 and 1),
choice (one of the offered windows, or an empty string), counter_window (ISO-8601 time with
UTC offset, or an empty string), notes (one short sentence quoting the person, in their language).

Requester call, rendered and spoken in requester.language, includes the recipient's captured answer verbatim:

text
You are an AI phone assistant calling {requester_first_name} back about {topic}.
Disclose immediately that you are an AI assistant.
Speak {requester_language} throughout.

Relay the recipient's answer exactly: {recipient_answer_summary}
If they chose a window, read it back and ask the requester to accept it.
If they proposed a different time or declined, relay that without negotiation.
Capture accept, decline, or next instruction. Do not commit to anything beyond relaying.

Record the outcome as JSON with exactly these fields: accepted (boolean),
agreed_window (one of the offered windows, or an empty string), notes (one short sentence).
Show full SKILL.md (324 more words)Show less

Structured Result

json
{
  "request_id": "relay-2026-09-08-unit-12",
  "relay": "agreed | needs_human",
  "agreed_window": "ISO-8601 or null",
  "calls_placed": "0, 1, or 2",
  "recipient_call": {
    "disposition": "completed | voicemail | no_answer | wrong_number | no_consent | low_confidence | schema_drift",
    "answer": {"understood": true, "consent_given": true, "wrong_person": false, "confidence": 0.92, "choice": "ISO or empty", "counter_window": "ISO or empty", "notes": "one sentence in recipient language"}
  },
  "requester_call": {"disposition": "completed | skipped | ...", "answer": {"accepted": true, "agreed_window": "ISO or empty", "notes": "..."}},
  "needs_human_reason": "string or null"
}

Fail closed: voicemail, no-answer, wrong person, missing consent, low-confidence answers, and schema drift all become needs_human with the reason filled in. The recipient answer is bound to a typed contract before the requester call is planned: understood, consent_given, wrong_person must be booleans and confidence a number in [0, 1] — a missing or mistyped field is schema drift, a missing or non-true consent_given refuses the second call, and confidence below 0.7 routes to a human. The requester answer is type-checked the same way (accepted boolean, agreed_window/notes strings). An agreed result is a relayed confirmation, not a contract; the requester confirms any commitment themselves on the second call.

Live Planning

Only after explicit user authorization and CALL-E authentication:

bash
python3 skills/language-bridge-call/scripts/relay_call.py --request <authorized-request.json> --execute --confirm-consent

The script runs calle call plan, then calle call run --plan-id --confirm-token, then polls calle call status --run-id until a terminal status, for the recipient call first and the requester call second. If run_call has an uncertain outcome, the script stops and prints the exact calle call recover --recovery-id <id> command instead of retrying.

Planning without execution is always available and is not a call:

bash
calle call plan --to-phone <E164_PHONE> --goal "<reviewed goal text>" --timezone America/Los_Angeles --language Spanish --region US

Do not run --execute unless the user separately confirms both numbers and the request text.

Cancellation And Idempotency

Idempotency key: language_bridge_call:{request_id}, enforced by the script on --execute. Before any call is placed it records status: started in ~/.cache/language-bridge-call/<sha256 of request_id>.json and marks it done with the final result afterwards; a re-run with the same request_id is refused (exit 2) with the prior status printed. If a run ends uncertainly, resolve it with the printed calle call recover command (or restart with a new request_id) rather than re-running; delete the state file only if you have verified no call was placed. Cancel before dial by not running --execute; there is no background job to cancel because the script does not schedule anything. Ambiguous outcomes route to a human rather than placing another call.

Safety Notes

Read references/safety.md and references/examples.md before live planning.

© CALLE-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (scripts, references, assets) in skills/language-bridge-call of CALLE-AI/awesome-phone-call-agents.

  • SKILL.md
  • assets/sample-relay-request.json
  • references/examples.md
  • references/safety.md
  • scripts/fixtures/relay_happy_path.json
  • scripts/relay_call.py
  • scripts/test_relay_call.py

Open the folder on GitHubat commit 38d4118

Compare with similar skills

Language Bridge Call next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Language Bridge Call compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Language Bridge Call this skillCALLE-AI/awesome-phone-call-agents107—~2.3kAutomated safety check: PassMIT
Memory Bridgeruvnet/ruflo74k—~601Automated safety check: NotesMIT
Cross Chain Relayer Auditsickn33/agentic-awesome-skills47k1 repos~1.4kAutomated safety check: PassMIT
Cowork To Code Bridgesickn33/agentic-awesome-skills47k1 repos~2.6kAutomated safety check: PassMIT
Relaysangrokjung/claude-forge852—~1.5kAutomated safety check: PassMIT
Relay Transportopenchamber/openchamber11k—~1.8kAutomated safety check: PassMIT

Similar skills

  • Memory Bridge

    ruvnet/ruflo

    Bridge Claude Code auto-memory into AgentDB with ONNX embeddings, deduplicate, and enable unified cross-project search

    74k GitHub stars~601 tokensUpdated yesterday
    AI & LLM EngineeringAuto-check: notes
  • Cross Chain Relayer Audit

    sickn33/agentic-awesome-skills

    Cross-chain relayer bridge audit register: message hash verifications, replay protection nonces, validator quorum, and withdrawal proofs.

    47k GitHub starsUsed in 1 repo~1.4k tokens
    SecurityAuto-check passed
  • Cowork To Code Bridge

    sickn33/agentic-awesome-skills

    Use an already-installed, independently verified cowork-to-code bridge to run narrowly approved actions on the user's own macOS, Linux, or WSL2 machine through a local file queue.

    47k GitHub starsUsed in 1 repo~2.6k tokens
    Auto-check passed
  • Relay

    sangrokjung/claude-forge

    Session relay — compresses what's been done so far (writes a one-liner for /compact) and generates a "next task" prompt (a baton) for the user to hand off.

    852 GitHub stars~1.5k tokensUpdated 1 mo ago
    Agent WorkflowsAuto-check passed
  • Relay Transport

    openchamber/openchamber

    A skill your agent uses when adding or changing OpenChamber WebSocket, SSE, streaming, realtime endpoints, shared UI sockets, runtime transport internals, private relay behavior, or files under the…

    11k GitHub stars~1.8k tokensUpdated today
    Backend & APIsAuto-check passed
  • Web3 Transaction Relayer Pool

    sickn33/agentic-awesome-skills

    Gasless transaction relayer node pool register: fee sponsorship limits, nonce synchronization, and balance replenishment alerts.

    47k GitHub starsUsed in 1 repo~1.3k tokens
    Business, Finance & HRAuto-check passed

More from CALLE-AI/awesome-phone-call-agents

All 101 skills in this repo
  • Accessible Outing Verifier

    CALLE-AI/awesome-phone-call-agents

    Demonstrates advisory accessibility-planning checks with offline fixtures and a proposed bounded CALL-E workflow; use for exploring unknown or qualified venue claims without making calls.

    107 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed
  • Ground Truth Gate

    CALLE-AI/awesome-phone-call-agents

    A skill your agent uses when an agent holds some evidence for a physical-world claim but the evidence is broader, narrower, or older than the exact question asked, and it must first decide whether a…

    107 GitHub stars~3.3k tokensUpdated yesterday
    Auto-check passed
  • Is It Accessible

    CALLE-AI/awesome-phone-call-agents

    Call a venue and ask the accessibility questions that matter to one specific person — step-free entry, hearing loop, guide dogs, quiet hours, changing places — then return a per-need verdict backed…

    107 GitHub stars~4.3k tokensUpdated yesterday
    Auto-check passed
  • Landmark Navigation Assist

    CALLE-AI/awesome-phone-call-agents

    Turns a pre-written, building-level location config into a CALL-E outbound phone-call task that guides a delivery driver through the last few hundred metres to a specific building using landmarks…

    107 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed
  • Research Gap Call Verifier

    CALLE-AI/awesome-phone-call-agents

    Turn cited business research into a bounded, approval-gated phone-call plan that asks only unresolved factual questions, then reconcile CALL-E-compatible results without treating voicemail, refusal…

    107 GitHub stars~1.7k tokensUpdated yesterday
    Auto-check passed
  • Structured Outcome Followup Call

    CALLE-AI/awesome-phone-call-agents

    Place a goal-driven CALL-E call that collects specific structured answers, score those answers against a deterministic rubric you supply, and conditionally trigger a follow-up action — all runnable…

    107 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed

Questions about Language Bridge Call

What does Language Bridge Call do?

Cross-language relay call skill for CALL-E - places one call in the recipient's language to capture a structured answer, then one call back to the requester in their language, and returns an agreed…. Language Bridge Call is an agent skill from CALLE-AI/awesome-phone-call-agents. Cross-language relay call skill for CALL-E - places one call in the recipient's language to capture a structured answer, then one call back to the requester in their language, and returns an agreed outcome or a fail-closed handoff.

How do I install Language Bridge Call in Claude Code?

Run `npx skills add CALLE-AI/awesome-phone-call-agents --skill language-bridge-call -a claude-code`. Or copy the skill folder (skills/language-bridge-call in CALLE-AI/awesome-phone-call-agents) into .claude/skills/language-bridge-call in your project. Claude Code loads it when a task matches its description.

How do I install Language Bridge Call in Codex?

Run `npx skills add CALLE-AI/awesome-phone-call-agents --skill language-bridge-call -a codex`. Or copy the skill folder (skills/language-bridge-call in CALLE-AI/awesome-phone-call-agents) into .agents/skills/language-bridge-call in your project. Codex loads it when a task matches its description.

Can I use Language Bridge Call in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add CALLE-AI/awesome-phone-call-agents --skill language-bridge-call -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/language-bridge-call, .gemini/skills/language-bridge-call, .github/skills/language-bridge-call and .opencode/skills/language-bridge-call in your project.

What does Language Bridge Call need to run?

Going by SKILL.md and its folder, Language Bridge Call needs Python for the scripts in its folder and the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Language Bridge Call access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Language Bridge Call safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Language Bridge Call use?

Language Bridge Call is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Language Bridge Call use?

About 2.3k tokens (SKILL.md is roughly 9.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 731 tokens, read only when the agent opens those files.

What are the alternatives to Language Bridge Call?

Skills that share tags, products or a category with Language Bridge Call: Memory Bridge (ruvnet/ruflo, 74k stars), Cross Chain Relayer Audit (sickn33/agentic-awesome-skills, 47k stars), Cowork To Code Bridge (sickn33/agentic-awesome-skills, 47k stars) and Relay (sangrokjung/claude-forge, 852 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Language Bridge Call?

CALLE-AI (a GitHub organization) maintains it in CALLE-AI/awesome-phone-call-agents, which has 107 GitHub stars. The repository holds 101 skills in this directory. The repository was last updated on October 10, 2026.

Source: CALLE-AI/awesome-phone-call-agents on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.