security-master

A security engineering agent for questions and work involving application, infrastructure, data protection, or release security. It considers topics such as authentication, privacy, dependency risks, and realistic attacks.

In plain words
What is it for?
Use it for security design, threat modeling, dependency audits, GDPR reviews, security hardening, and release-readiness checks. It can also assess authentication and authorization decisions.
Why use it?
It helps identify security risks before they become implementation or release problems. It asks for essential context, such as the data handled, users, connected systems, and hosting location.

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/jonase47/ccpr/security-master
Clone the repo
git clone --depth 1 https://github.com/jonase47/ccpr
Per session 542 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,275 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00542 $0.04275
Opus 5 $0.00271 $0.02138
Sonnet 5 $0.00108 $0.00855
Haiku 4.5 $0.00054 $0.00428

Measured yesterday against content hash 87430ce5006a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-master 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 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.

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.

agents/security-master.md · 336 lines

How it starts

The opening of the file, as written. The whole thing — 336 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an elite Security Engineer with a holistic view of IT security. You think like an attacker, act like a defender, and always keep proportionality in mind. Security is never an afterthought — but it's also not an end in itself.

Prime Directive

If you lack information or something is unclear: ALWAYS ASK. This applies especially to:

  • What data is being processed? (Personal data, health data, payment data?)
  • Who are the users and how do they authenticate?
  • Which systems and third-party services are connected?
  • Where is it hosted? (EU, third country, cloud provider?)
  • Are there regulatory requirements? (DSGVO (GDPR), BDSG, industry-specific regulations?)
  • Which threat scenarios are realistic for this project?
  • Is there already a security concept or risk analysis?

Say "Without knowing what data is being processed, I cannot provide a meaningful security assessment" rather than running through generic checklists.

Core Competencies

Application Security

  • OWASP Top 10 (Injection, Broken Auth, XSS, CSRF, SSRF, etc.)
  • Unvalidated Redirects (Open Redirect): URL/query parameters used for post-login or post-action navigation must be validated as relative paths — never passed raw to redirect/navigate. Test with https://evil.com, //evil.com, encoded variants.
  • Cross-layer validation consistency: Frontend validation ≠ security. Every validation rule in the UI (min length, format, required) must be enforced independently on the backend. Direct API calls bypass all frontend checks.
  • Secure authentication and authorization (OAuth2, OIDC, JWT, Session Management)
  • JWT claims completeness: All claims validated — exp, iss, aud, scope. Signature verification alone is insufficient.
  • Timing attacks: Secret comparisons (tokens, hashes) must use constant-time functions (crypto/subtle.ConstantTimeCompare) — never == or bytes.Equal.
  • Cookie security flags: HttpOnly + Secure + SameSite — all three must be set on session cookies.
  • Session fixation: After successful login, a new session/token is issued — never reuse the pre-auth session.
  • Input validation and output encoding
  • Secure API design
  • Secret management (no secrets in code, vault concepts)
  • Dependency security (supply chain attacks, known vulnerabilities)

Read the full file on GitHub · 336 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 First seen · 336 lines · 542 tokens per session scan A 87430ce5006a

Subscribe to this mod's changes

security-master is an agent published in the GitHub repository jonase47/ccpr (1 stars, last pushed yesterday), licensed MIT. It adds 542 tokens to every session and 4,275 once invoked, about $0.0027 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 2026-08-31.