kyde score

com.kydePublished by kyde.comOfficial publisherv0.8.24Updated Nov 7, 2026

Agent observability via OpenTelemetry: drift, loop and hijack alerts, e.g. after a prompt injection.

VerifiedStreamable HTTPWeb executableSecurity & MonitoringAI & ML

Overview

AI-generated overview

Remote MCP server that scores AI agent behavioral drift from OpenTelemetry data and alerts on band changes.

What you can do
  • Create and manage a scoring key, optionally with an HTTPS webhook for alerts
  • Read current 0-100 drift scores per agent, with a learning status until a 30-run baseline freezes
  • Set or change the webhook that receives HMAC-signed POSTs on band changes and rebaselines
  • Claim the key with a human email via double opt-in so a person gets the dashboard
  • Rebaseline one agent to relearn from its next 30 runs
  • Revoke the key and permanently delete all data stored under it
When to use it
Useful when you run AI agents instrumented with OpenTelemetry and want an outside signal that their behavior has drifted, looped, or been hijacked, for example after a prompt injection. It fits teams that want band-change alerts delivered to a human rather than polling scores.
Requirements
A remote endpoint at score.kyde.com; no local package. A scoring key, created with the create_key tool or supplied as the x-kyde-key header or the key argument. Agents must send OTLP/HTTP telemetry (protobuf or JSON) to the endpoint with the x-kyde-key header. A human email is needed to claim the key; an HTTPS webhook URL is optional.
Before you install
The scoring key is a secret: anyone holding it can read scores and, until the key is claimed, change the webhook or rebaseline. Claiming the key by email hands control of rebaseline, webhook changes and revocation to that person. Revoking deletes all stored data permanently and requires confirm=true. Agent telemetry, including service names and run data, is sent to a third-party service.

Installation

In SourceWeft

  1. Open kyde score in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/com.kyde%2Fscore/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "kyde-score": {
      "type": "http",
      "url": "https://score.kyde.com/mcp"
    }
  }
}

Tools

6

create_key

Create scoring key

Create a scoring key (no signup). One key covers all agents, identified by OTel service.name. webhook_url (https) is optional but recommended: it receives an HMAC-signed POST when any agent's score changes band.

Parameterssourcewebhook_url

get_scores

Get drift scores

Current drift score per agent under the key: 0-39 deviating, 40-69 watch, 70-100 normal; status 'learning' until the 30-run baseline freezes.

Parameterskey

set_webhook

Set alert webhook

Add or change the https webhook. It is POSTed on every band change and rebaseline, HMAC-SHA256-signed with the scoring key (header x-kyde-signature: sha256=<hex> over the raw body).

Parameterskeywebhook_urlrequired

claim_key

Attach a human to the key

Attach your human's email to the key (double opt-in: Kyde mails a confirmation link). Once confirmed, the key is claimed: the human gets the management dashboard, and rebaseline, webhook changes and revocation work only for them — no agent, including you, can reset its own score. Ask the human once for the address; never invent or guess one; skip when get_scores already shows claimed: true.

Parameterskeyemailrequiredproduct_updates

rebaseline

Reset agent baseline

Forget one agent's learned baseline and score and relearn from its next 30 runs. For intended behavior changes (new model, new task) that left the agent permanently deviating. The reset is permanently visible: records and alarms stay stored, the agent carries rebaselined_at, and the reset itself is recorded as a rebaseline alarm naming the score/band it erased. Requires confirm=true and the human's explicit go-ahead.

Parameterskeyconfirmrequiredservice_namerequired

revoke_key

Revoke key and delete data

Revoke the key and permanently delete all data stored under it. Requires confirm=true.

Parameterskeyconfirmrequired

Version history

6
  1. v0.8.24LatestNov 7, 2026
  2. v0.8.17Nov 6, 2026
  3. v0.8.8Nov 6, 2026
  4. v0.8.6Nov 6, 2026
  5. v0.5.2Nov 4, 2026
  6. v0.2.6Nov 3, 2026

Score

72% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts

    Publish an icon in the registry entry.

  • Documentation15 pts

    Declare a GitHub repository with a README, or serve resource_documentation in the OAuth protected-resource metadata.

  • LicenseNot applicable

    A hosted service is used under its provider's terms, so no license is expected here. This check is left out of the percentage.