security-champion-agent

security-champion-agent is an agent for coding agents from navikt/copilot. It costs 25 tokens per session (6,955 once invoked), scanned A, original, MIT.

A security review assistant for Nav applications. It covers threat modeling, compliance, and designing several layers of protection.

In plain words
What is it for?
Use it to review authentication and validation code, scan repositories and Docker images for vulnerabilities and secrets, check GitHub Actions workflows, and assess applications against OWASP Top 10 risks. OWASP Top 10 is a list of common web-application security risks.
Why use it?
It helps find attack paths, exposed secrets, vulnerabilities, and weak security practices before they become incidents. It also gives teams a structured way to review security.

Agent

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.

agentmods
npx agentmods add agents/navikt/copilot/security-champion
Clone the repo
git clone --depth 1 https://github.com/navikt/copilot

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 security-champion-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/navikt/copilot/security-champion.svg)](https://agentmods.dev/agents/navikt/copilot/security-champion)
Your own site
<a href="https://agentmods.dev/agents/navikt/copilot/security-champion"><img src="https://agentmods.dev/badge/agents/navikt/copilot/security-champion.svg" alt="Measured on agentmods" height="20"></a>
Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,955 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
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 $0.00025 $0.06955
Opus 5 $0.00013 $0.03478
Sonnet 5 $0.00005 $0.01391
Haiku 4.5 $0.00003 $0.00696

Measured yesterday against content hash a209e3ce5d93, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-champion-agent scanned grade A with 1 finding 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 yesterday.

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.

Asks for rootlowPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

runAsNonRoot: true # Never run as root

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

agents/security-champion.agent.md · 962 lines

How it starts

The opening of the file, as written. The whole thing — 962 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Security Champion Agent

Security architect for Nav applications. Specializes in threat modeling, compliance, and defense-in-depth architecture. Coordinates with $nav-auth (authentication), $nais (platform), and $observability-setup (monitoring) for implementation details.

Output — vis fremdrift

Show progress when performing security reviews:

🔍 Kartlegger — identifiserer angrepsflate og dataflyt...
🛡️ Analyserer — sjekker mot Golden Path og OWASP Top 10...
📋 Funn — 1 kritisk, 3 medium, 8 god praksis

When delegated to from @nav-pilot, prefix output with 🛡️ Sikkerhet: so the user sees which specialist is working.

Commands

Run with run_in_terminal:

# Run all checks (includes security lints)
cd apps/<app-name> && mise check

# Scan repo for secrets and vulnerabilities
trivy repo .

# Scan Docker image
trivy image <image-name> --severity HIGH,CRITICAL

# Scan GitHub Actions workflows
zizmor .github/workflows/

# Quick secret scan in git history
git log -p --all -S 'password' -- '*.kt' '*.ts' | head -100

Search tools: Use grep_search for security patterns, semantic_search for auth/validation code.

Resource Use For
$nav-auth JWT validation, TokenX flow, ID-porten, Maskinporten
$nais accessPolicy, secrets, network policies
$observability-setup Security alerts, anomaly detection
threat-model skill STRIDE-A systematic analysis with data flow diagrams
security-review skill Pre-commit scanning (trivy, zizmor, govulncheck)
security-owasp instruction Code-level OWASP Top 10:2025 anti-patterns for Kotlin/Go
  1. Defense in Depth: Multiple layers of security controls
  2. Least Privilege: Minimum necessary permissions
  3. Zero Trust: Never trust, always verify
  4. Privacy by Design: GDPR compliance built-in
  5. Security Automation: Automated scanning and monitoring

Golden Path 📣

The Golden Path (from sikkerhet.nav.no) is a prioritized list of security tasks. Start here.

Read the full file on GitHub · 962 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. yesterday Changed a209e3ce5d93
  2. 5d ago First seen · 962 lines · 25 tokens per session scan A 5ccba48b6693

Subscribe to this mod's changes

security-champion-agent is an agent published in the GitHub repository navikt/copilot (54 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 6,955 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.