migrate-from-retell
Skill0 starsChanged 57 days agoMITView on GitHub
Import and validate Retell AI agents in ThunderPhone. Use when someone asks to migrate from Retell, create a Retell agent-import job, review response-engine or flow mappings, identify manual tool and number work, commit selected items, or run parity tests before cutover.
- Reads credentials
- Sends data out
In this repository21 other files it gives its agents.
- AGENTS.mdAGENTS.md
- add-knowledgeSkill
- build-tool-integrationSkill
- create-agentSkill
- embed-web-widgetSkill
- get-phone-numberSkill
- handle-inbound-callsSkill
- iterate-on-agentSkill
- migrate-from-blandSkill
- migrate-from-elevenlabsSkill
- migrate-from-vapiSkill
- place-outbound-callsSkill
- run-campaignSkill
- setup-api-keySkill
- setup-webhooksSkill
- test-agentSkill
- thunderphone-mcp-setupSkill
- thunderphone-prompt-builderSkill
- use-with-livekitSkill
- use-with-openai-realtimeSkill
- use-with-pipecatSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 16
Choose
api_keyorpastemode: Use direct access only with an authorized source credential. - Very specific: line 54
Poll and edit review items: Use GET /v1/agent-imports/{public_id}, then PATCH /v1/agent-imports/{public_id}/items/{item_id} for selection, …
- Distinctive: line 84
Use import agents and the agent-import API.
- Distinctive: line 86
Do not claim feature parity from a successful commit alone.
- Distinctive: line 87
Keep traffic changes, number moves, and credential revocation outside the review import.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Migrate from Retell
- Inventory the source. Export agents, response engines or flow states, numbers, tools, knowledge, webhooks, credentials, traffic, and test cases. Record the source revision and unresolved ownership before any import.
- Choose
api_keyorpastemode. Use direct access only with an authorized source credential. Keep that credential in the protected request flow; never put it in logs, prompts, or committed JSON. - Create the job with
POST /v1/agent-imports,source: retell, and the selected auth mode.
{
"source": "retell",
"auth_mode": "paste",
"payload": "${EXPORTED_AGENT_JSON}"
}curl --fail-with-body -X POST https://api.thunderphone.com/v1/agent-imports \
-H "Authorization: Bearer $THUNDERPHONE_API_KEY" -H "Content-Type: application/json" \
-H "X-ThunderPhone-Client: skills/migrate-from-retell@1.0" --data-binary @retell-import.jsonimport json, os, requests
with open("retell-import.json", encoding="utf-8") as f: payload = json.load(f)
r = requests.post("https://api.thunderphone.com/v1/agent-imports", json=payload,
headers={"Authorization": f"Bearer {os.environ['THUNDERPHONE_API_KEY']}",
"X-ThunderPhone-Client": "skills/migrate-from-retell@1.0"}, timeout=60)
r.raise_for_status(); print(r.json())import { readFile } from "node:fs/promises";
const r = await fetch("https://api.thunderphone.com/v1/agent-imports", { method: "POST",
headers: { Authorization: `Bearer ${process.env.THUNDERPHONE_API_KEY}`,
"Content-Type": "application/json", "X-ThunderPhone-Client": "skills/migrate-from-retell@1.0" },
body: await readFile("retell-import.json", "utf8") });
if (!r.ok) throw new Error(await r.text()); console.log(await r.json());- Poll and edit review items. Use
GET /v1/agent-imports/{public_id}, thenPATCH /v1/agent-imports/{public_id}/items/{item_id}for selection, name, voice, product, or prompt choice. - Check mapping and gaps. General/global prompts, flow state content, supported tools, transfer/end/DTMF, voice, language, model, knowledge, voicemail, silence, keywords, and duration can map. Numbers, webhooks, credentials, custom engines, SMS tools, and vendor MCP settings need manual review. See references/retell-mapping.md.
- Commit reviewed items with
POST /v1/agent-imports/{public_id}/commit; poll to terminal and record the created IDs. Delete an abandoned uncommitted job explicitly. - Run matched side-by-side tests with
test-agent. Verify every flow transition, DTMF/tool action, knowledge result, transfer, language, audio, and end condition. Traffic cutover and source retirement are separate, approval-gated work with rollback. - Use live comparison data from
https://thunderphone.com/data/voice-ai-price-index.json; never invent or freeze a competitor rate in the skill.
Errors
| Symptom | Cause | Fix |
| Response engine is incomplete | Custom engine or flow construct is unsupported | Preserve the source and rewrite the behavior as reviewed prompt/tool logic. |
| DTMF or tool action differs | Vendor action schema did not map exactly | Rebuild and test the action with explicit inputs and success evidence. |
| Import auth fails | Retell access or auth mode is invalid | Use a correctly scoped credential or permitted paste payload. |
| Tests are not comparable | Source and destination scenarios/config differ | Freeze the case set and record both revisions and environments. |
Source and safety rules
- Use import agents and the agent-import API.
- Protect source credentials and preserve unmodified source assets until cutover is verified.
- Do not claim feature parity from a successful commit alone.
- Keep traffic changes, number moves, and credential revocation outside the review import.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

