When was castmyagent.ai last audited?

Last checked 2026-10-27; wmcp.sh re-checks it on a schedule for tool drift and rug-pulls.

castmyagent.ai

Item: castmyagent.ai
Rating: 45
Author: wmcp.sh

https://castmyagent.ai/api/mcp

45/100 · MCP Trust Grade · checked 2d ago · MCP 2025-06-18

Is the castmyagent.ai MCP server safe to use?

Independent trust grade F (45/100). Tool "get_persona" description contains prompt-injection / hidden-instruction markup. wmcp.sh continuously watches castmyagent.ai for tool drift and rug-pulls. The grade is free and identical whether or not the operator pays.

✓ Watched since 2026-10-23 — behavioral baseline locked, no drift for 3 days. We re-check this server's tool surface on a schedule; if it adds, removes, or silently rewrites a tool (rug-pull), we record it.

What it offers — 3 tools · Commerce

list_personas

List free CastMyAgent persona packs. Optionally filter by department. Returns slug, name, role, department, a short summary, and the roster URL.

get_persona

Load one free persona pack by slug. Returns the production system prompt, the character brief, and the style configuration. The pack is free and the t

search_personas

Find free persona packs by keyword. Matches the role and specialties.

Spec conformance20%100

✓ initialize → protocolVersion 2025-06-18
✓ initialize HTTP 200
✓ unknown method → -32601
✓ tools/list → 3 tools
✓ every tool has name + inputSchema

Security (OWASP MCP)30%15

3 tools scanned
1 injection-markup hits
no secret-path refs

Reliability / performance20%70

single-probe latency 1774ms (PRELIMINARY — continuous uptime/p95 from proxy telemetry lands in v2)

Tool hygiene15%80

100% of tools have typed inputSchema
0/3 declare outputSchema
no unannotated destructive tools

Transparency / provenance15%70

HTTPS ✓
advertises protocol 2025-06-18

Observed behavior

No proxied traffic observed for this host yet. Connect it at /connect and its grade gains a measured Reliability score + per-tool behavioral evidence — the half a static scan can't produce.

Findings

FAILMCP01 Tool "get_persona" description contains prompt-injection / hidden-instruction markup.

Developers see this F before they connect to castmyagent.ai.

Clear it: a Deep Audit shows exactly what to fix, and Monitoring proves the fix to everyone who checks your grade. Drop your email and we'll send this server's full findings + the fix path. The grade itself stays free.

Grade another server

We re-grade castmyagent.ai on a schedule and alert your Slack/webhook the moment its tools change or its grade drops — rug-pull insurance for the connection.

Run castmyagent.ai? Claim it (free) to get drift alerts and show an independently-verified trust badge. The grade stays free — claiming just ties it to you.

Share this report card

A 1200×630 card with the grade + audit — drop it in a post, Slack, or your repo.

MCP Trust report card — castmyagent.ai grade F

Share on X Open card image

https://wmcp.sh/mcp/grade/castmyagent.ai

Embed this grade

A live badge — it re-verifies itself and shows current stability. Static scorecards can't. Paste it in your README or site to show users you're independently audited.

Markdown (for your README)

[![MCP Trust Grade F](https://wmcp.sh/mcp/grade/castmyagent.ai/badge.svg)](https://wmcp.sh/mcp/grade/castmyagent.ai)

HTML

<a href="https://wmcp.sh/mcp/grade/castmyagent.ai"><img src="https://wmcp.sh/mcp/grade/castmyagent.ai/badge.svg" alt="MCP Trust Grade F · wmcp.sh"></a>

Agents: check this before connecting

Add the wmcp.sh trust oracle as an MCP server and call grade_mcp_server / check_mcp_drift in your agent's pre-connection gate:

https://wmcp.sh/mcp/trust

How this grade is computed. An open, independent rubric — Spec conformance (20%), Security mapped to the OWASP MCP Top 10 (30%), Reliability (20%), Tool hygiene (15%), Transparency (15%) — run by connecting to the server and inspecting its real MCP surface. The grade is free and identical whether or not the operator pays. v1 uses static + spec signals from a single connection; continuous uptime, real latency, and annotation-truthing (declared readOnly vs observed behavior) layer on via the wmcp.sh proxy.