rtcfSAFE
Prompt improver MCP server (hosted). Turns rough requests into sharp Role/Task/Context/Format prompts. Thai + English.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Turn rough requests into sharp, structured prompts, right inside Claude, ChatGPT, or any MCP client.
RTCF rewrites a messy request into the four parts every strong prompt needs: Role, Task, Context, Format. Works in Thai and English. Free tier included (5 credits/day, no sign-up).
- Website: https://rtcf.happlen.com
- Setup guide: https://rtcf.happlen.com/docs
- By Happlen
Endpoint
This is a hosted (remote) MCP server. No installation, no OAuth setup:
https://rtcf.happlen.com/api/mcp
Transport: Streamable HTTP. On a paid plan, append your key: https://rtcf.happlen.com/api/mcp?key=rk_live_...
Tools
Quick setup
Claude (web / desktop / mobile)
- Settings -> Connectors -> Add custom connector
- Paste
https://rtcf.happlen.com/api/mcp(leave OAuth fields empty) - In a chat, enable the connector and say: "Improve this prompt with RTCF, then answer it."
ChatGPT (paid plans, Developer mode)
- Settings -> Apps -> Advanced -> enable Developer mode
- Add a custom connector with the same URL
- Ask ChatGPT to use RTCF
Claude Code
claude mcp add --transport http rtcf https://rtcf.happlen.com/api/mcp
REST API
curl -X POST https://rtcf.happlen.com/api/v1/convert \
-H "Authorization: Bearer rk_live_..." \
-H "Content-Type: application/json" \
-d '{"prompt": "help me write an email to my professor"}'Full API docs: https://rtcf.happlen.com/docs
ภาษาไทย
RTCF คือผู้ช่วยปรับพรอมต์สําหรับคนที่ใช้ AI แล้วไม่รู้จะพิมพ์อะไรดี วางคําขอคร่าว ๆ แล้วรับพรอมต์โครงสร้าง Role, Task, Context, Format กลับไปใช้ได้ทันที ใช้ฟรีวันละ 5 เครดิต ไม่ต้องสมัครสมาชิก ใช้ได้ทั้งบนเว็บ
1967d730e41dOBSERVED · 2026-10-11Trust audit
SAFEgrade B · trust 87/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (3)
streamable-http
Gates applied: no_behavioural_pass, no_license.
1967d730e41dfull audit observations/trust-audit/mcp-server/hengkp__rtcf-mcp.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-11 | 1967d730e41d | SAFE | B | 87 | first audit |
Questions
What is the rtcf MCP server?
Prompt improver MCP server (hosted). Turns rough requests into sharp Role/Task/Context/Format prompts. Thai + English.
Is rtcf safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (87/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does rtcf need?
No credential environment variables were found in its source, so it appears to need none.
How does rtcf run?
It speaks streamable-http, so it runs as a service you connect to over the network.
How current is this page?
The grade is for one exact copy of the source (1967d730e41d), read on 2026-10-11. The repository is watched and re-audited when it changes.