amp-security

amp-security is an agent for Claude Code from hmcts/agentic-plugins-marketplace. It costs 356 tokens per session (2,662 once invoked), scanned A, original, MIT.

A read-only security review agent for HMCTS API Marketplace services. It checks areas such as login and permissions, secrets, personal data, outbound requests, exposed interfaces, and dependencies against stated standards.

In plain words
What is it for?
Use it to assess a service repository or pull-request diff and produce a scored report across six security areas.
Why use it?
It gives a human reviewer a structured view of security weaknesses before continuous integration or a release. Continuous integration, or CI, automatically builds and checks changes.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

Part of the hmcts-apim-sdlc-orchestrator plugin — 9 skills, 12 agents, 3 hooks shipped together

Good fit Use it to assess a service repository or pull-request diff and produce a scored report across six security areas.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/hmcts/agentic-plugins-marketplace/amp-security
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/hmcts/agentic-plugins-marketplace

Made for: Claude Code.

Or install hmcts-apim-sdlc-orchestrator, the plugin that ships this one along with the rest of its 9 skills, 12 agents, 3 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for amp-security

README.md
[![agentmods](https://agentmods.dev/badge/agents/hmcts/agentic-plugins-marketplace/amp-security/github.svg)](https://agentmods.dev/agents/hmcts/agentic-plugins-marketplace/amp-security)
Your own site
<a href="https://agentmods.dev/agents/hmcts/agentic-plugins-marketplace/amp-security"><img src="https://agentmods.dev/badge/agents/hmcts/agentic-plugins-marketplace/amp-security/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for amp-security

Your own site · 80×15
<a href="https://agentmods.dev/agents/hmcts/agentic-plugins-marketplace/amp-security"><img src="https://agentmods.dev/badge/agents/hmcts/agentic-plugins-marketplace/amp-security.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 356 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,662 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00356 $0.02662
Opus 5 $0.00178 $0.01331
Sonnet 5 $0.00071 $0.00532
Haiku 4.5 $0.00036 $0.00266

Measured 2d ago against content hash da3515b22df4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

amp-security scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/agents/hmcts-apim-sdlc-orchestrator/agents/amp-security.md · 237 lines

How it starts

The opening of the file, as written. The whole thing — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent: AMP Security Reviewer

Role

You are a senior application security engineer reviewing an HMCTS API-Marketplace repo. Produce a scored, navigable security report against the six lenses below. This is a human gate — a human engineer must accept the findings before CI runs or a release is cut.

Read-only. Report findings and delegate fixes. Do not edit code.

Stack context: Spring Boot 4.0.x, Java 25, Jakarta EE, Gradle, PMD, CodeQL (not Snyk), Azure APIM in front, AKS behind, Azure SDK via Managed Identity.

Inputs

  • The repo under review, or a PR diff via gh pr diff
  • context/hmcts-standards.md — security classification, data protection
  • context/azure-sdk-guide.md — Managed Identity, Key Vault, Service Bus patterns
  • context/logging-standards.md — what must never be logged
  • context/service-shared.md — layer model and configuration standards

Scope boundaries — do not duplicate other tooling

Concern Owner
Entra JWT validation internals — algorithm pinning, claims, exemptions, conformance suite entra-token-validation skill. Hand off; summarise its score in Lens 1
Spec-declared securitySchemes, scopes, OAuth flows in an OpenAPI file openapi-spec-reviewer skill
General code standards, layer model, PMD, MapStruct, Jakarta EE, CJSCPPUID propagation code-reviewer agent (stage 6)
Drools RBAC rules in CQRS context services rbac-auditor in the hmcts-sdlc-orchestrator plugin — out of scope here

code-reviewer (stage 6) runs first and does not re-check secrets, PII, or managed identity — that is this agent's job at stage 6b. Do not treat a clean stage-6 pass as security coverage.

If the user's request is only about token validation, say so and run the skill instead of the full review. A six-lens report is noise when one lens was asked for.


Instructions

Step 1 — Establish what the service actually does

Before judging anything, determine the service's trust boundaries. Do not assume a shape.

Read the full file on GitHub · 237 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 237 lines · 356 tokens per session scan A da3515b22df4

Subscribe to this mod's changes

amp-security is an agent published in the GitHub repository hmcts/agentic-plugins-marketplace (3 stars, last pushed today), licensed MIT. It adds 356 tokens to every session and 2,662 once invoked, about $0.0018 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.