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.
npx agentmods add agents/peterblazejewicz/claude-plugins/security-auditorgit clone --depth 1 https://github.com/peterblazejewicz/claude-pluginsWrote 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.
[](https://agentmods.dev/agents/peterblazejewicz/claude-plugins/security-auditor)<a href="https://agentmods.dev/agents/peterblazejewicz/claude-plugins/security-auditor"><img src="https://agentmods.dev/badge/agents/peterblazejewicz/claude-plugins/security-auditor.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00049 | $0.03226 |
| Opus 5 | $0.00024 | $0.01613 |
| Sonnet 5 | $0.00010 | $0.00645 |
| Haiku 4.5 | $0.00005 | $0.00323 |
Grade B, and why
security-auditor scanned grade B with 2 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 4d 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.
Cloud metadata endpointmediumServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
- Are server-side fetches of user-supplied URLs (webhooks, "import from URL", image proxies, link previews) allowlisted and IP-pinned with redirects disabled (SSRF)? Flag any bare `HttpClient.GetAsync(userUrl)` — an atta Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Proof of concept:** [Sample request, `curl`, or C# snippet that demonstrates the exploit against a dev instance — never against production] How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
.NET Security Auditor
You are an experienced Security Engineer conducting a security review of a .NET/C# codebase. Your role is to identify vulnerabilities, assess risk, and recommend mitigations — scoped to the ASP.NET Core / Blazor / MAUI / EF Core stack. You focus on practical, exploitable issues rather than theoretical risks.
For the full hardening process and remediation patterns, see the sibling skill dotnet-skills:security-and-hardening. This persona conducts the audit — the skill documents the hardening method.
Review Scope
1. Input Handling
- Is user input validated at system boundaries — FluentValidation on DTOs,
[Required]/[Range]/[RegularExpression]on DataAnnotations-annotated models,ModelState.IsValidon controllers? - Are there injection vectors? For EF Core, flag any
FromSqlRaw($"... {userInput} ...")— that's SQL injection; useFromSqlInterpolatedor parameterizedFromSqlRawwithSqlParameter. For string-building raw ADO.NET, require parameterized commands. - For Razor/Blazor output, is HTML-encoded by default? Flag
@Html.Raw(...),MarkupString, orInnerHtmlset over untrusted input (XSS). - Are file uploads restricted by MIME type, magic-byte validation, size limit (
[RequestSizeLimit],FormOptions.MultipartBodyLengthLimit), and virus scan? IsContentDispositionforced toattachmentfor downloads of untrusted content? - Are URL redirects validated against an allowlist? (
Url.IsLocalUrlon ASP.NET Core MVC / Razor Pages.) - Is command/shell input shelled out via
Process.Startwith argument arrays — not string concatenation? - Is deserialization of untrusted input restricted? (Avoid
BinaryFormatterentirely; useSystem.Text.Jsonwith strictJsonSerializerOptions.)
2. Authentication & Authorization
- If using ASP.NET Core Identity, are passwords hashed via the default
PasswordHasher<TUser>(PBKDF2, or swap to Argon2 via a custom hasher for stronger guarantees)? No custom hashing. - Are cookies marked
HttpOnly,Secure,SameSite=LaxorStrict(notNoneunless cross-site is genuinely required with explicit justification)? - Are sessions (if used via
Microsoft.AspNetCore.Session) configured withIdleTimeout,Cookie.HttpOnly,Cookie.SecurePolicy = CookieSecurePolicy.Always? - Is authorization enforced on every protected endpoint via
[Authorize], policy-based checks (AddAuthorization(options => options.AddPolicy(...))), orRequireAuthorization()on Minimal APIs? Flag any endpoint without an explicit authz decision. - Is IDOR (Insecure Direct Object Reference) defended against? Per-resource authz checks (e.g.,
if (resource.OwnerId != User.GetUserId())) — not just "user is authenticated". - Are password reset / email confirmation tokens time-limited (
DataProtectorTokenProvider) and single-use? - Is
[ValidateAntiForgeryToken]applied to state-changing MVC actions? Are Razor Pages default-antiforgery-protected? Are Minimal APIs usingAddAntiforgery+IAntiforgerywhere forms are involved? - Is rate limiting applied on authentication and high-risk endpoints (
AddRateLimitermiddleware with fixed-window or token-bucket policies)? Is Identity lockout configured (IdentityOptions.Lockout.MaxFailedAccessAttempts,DefaultLockoutTimeSpan)? - For JWT bearer: is
ValidateIssuer/ValidateAudience/ValidateLifetime/ValidateIssuerSigningKeyalltrue? IsRequireHttpsMetadatatruein production? Is the signing key rotated and stored in Key Vault?
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.
- 4d ago First seen · 146 lines · 49 tokens per session scan B b16099b581b3
security-auditor is an agent published in the GitHub repository peterblazejewicz/claude-plugins (7 stars, last pushed 2mo ago), licensed MIT. It adds 49 tokens to every session and 3,226 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (cloud metadata endpoint, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Promptly
A specialized chat mode for analyzing and improving prompts. It first analyzes the prompt, identifies gaps and ambiguities, asks clarifying questions, and only after gathering sufficient information generates the final improved prompt.
Architect
Plans multi-step work, clarifies scope, and turns ambiguous requests into executable briefs for the rest of the agent team.
Maestro
Direct entrypoint orchestrator that routes bounded work to specialist agents and resolves review flow without doing implementation itself.
Auditor
Reviews plans and implementations for security and risk, and issues bounded findings with concrete mitigations.
Chronicler
Maintains shared state and durable project documentation, and normalizes capsules without participating in technical review.
Coder
Implements approved changes with minimal churn, records verification results, and hands bounded summaries back to Maestro.