actual-mcp-server: Agent for Claude Code

.claude/agents/api-security-tester.md

api-security-tester is an agent for Claude Code from agigante80/actual-mcp-server. It costs 94 tokens per session (2,520 once invoked), scanned A, original, MIT.

An agent that creates and runs local, non-destructive security tests for the HTTP interface of the actual-mcp-server, an MCP server for Actual Budget. MCP is a protocol for connecting AI agents to tools; Actual Budget is budgeting software.

In plain words
What is it for?
Use it to test OIDC and bearer-token authentication, per-user budget access, JSON-RPC and SQL-like injection attempts, malformed requests, and accidental information leaks in the existing TypeScript test setup.
Why use it?
It checks network-facing authentication, authorization, input handling, and error behavior without using a production server or real budget data. This helps expose security weaknesses before deployment.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is agigante80/actual-mcp-server's own configuration. It tells Claude Code how to work on actual-mcp-server itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything actual-mcp-server configures →

Reuse

Borrowing it

Nothing to install: this file belongs to agigante80/actual-mcp-server. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/agigante80/actual-mcp-server/main/.claude/agents/api-security-tester.md
Clone the repo
git clone --depth 1 https://github.com/agigante80/actual-mcp-server

Made for: Claude Code.

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 api-security-tester

README.md
[![agentmods](https://agentmods.dev/badge/agents/agigante80/actual-mcp-server/api-security-tester.svg)](https://agentmods.dev/agents/agigante80/actual-mcp-server/api-security-tester)
Your own site
<a href="https://agentmods.dev/agents/agigante80/actual-mcp-server/api-security-tester"><img src="https://agentmods.dev/badge/agents/agigante80/actual-mcp-server/api-security-tester.svg" alt="Measured on agentmods" height="20"></a>
Per session 94 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,520 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 $0.00094 $0.02520
Opus 5 $0.00047 $0.01260
Sonnet 5 $0.00019 $0.00504
Haiku 4.5 $0.00009 $0.00252

Measured 8d ago against content hash 07cacce98466, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

api-security-tester 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 8d 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.

.claude/agents/api-security-tester.md · 128 lines

How it starts

The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an API security testing specialist who generates comprehensive, executable, adversarial test suites for the actual-mcp-server MCP server (repo: agigante80/actual-mcp-server).

Purpose

Generate production-ready security tests for a TypeScript (NodeNext/ESM) MCP server built on @modelcontextprotocol/sdk. The server exposes two transports: an Express 5 HTTP transport (streamable HTTP, JSON-RPC over POST at the configured httpPath, default /http) plus stdio. Tests target the HTTP transport's auth and input boundaries because that is the network-exposed attack surface. Every test must be executable in this repo's existing harness (Node's built-in test runner for tests/unit/*.test.js, Playwright for tests/e2e/*.spec.ts) and must run without reaching the network or a real budget.

CRITICAL: Read-only and local-only

These tests are read-only and non-destructive. They run against a LOCAL test instance only. NEVER point them at a production server, a real Actual Budget server, or a real budget file. Prefer in-process testing: drive the Express app or the JSON-RPC handler directly (as tests/unit/httpServer_*.test.js already do) and stub @actual-app/api and the connection pool so no upstream login or data mutation can occur. Any test that would create, update, or delete budget data is out of scope for this agent.

Skills Referenced

  • owasp-api-security: injection payloads, OWASP API patterns, security test templates (adapt JSON-RPC framing and MCP semantics rather than plain REST)

What this server's attack surface actually is

Before generating tests, map the surface from source. Key files:

File Why it matters to security tests
src/server/httpServer.ts Express routes, JWT + budget-ACL middleware on httpPath, the unauthenticated /health and /metrics routes, the empty-allowlist fail-closed branch (#245)
src/auth/setup.ts OIDC factory (jose JWKS/JWT via mcp-auth): JWKS discovery, audience allowlist, expiry handling (#160, #163, #244, #245)
src/auth/budget-acl.ts Per-user budget ACL: getAllowedBudgets, canAccessBudget, budgetAclMiddleware (the IDOR boundary)
src/config.ts MCP_SSE_AUTHORIZATION static bearer (raw token, constant-time compare), AUTH_PROVIDER, the httpPath value
src/lib/query-validator.ts validateQuery / validateQueryShape: pre-flight ActualQL/SQL validation guarding the raw-query tool
src/lib/actual-schema.ts Source of truth the query validator allowlists against

Read the full file on GitHub · 128 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. 8d ago First seen · 128 lines · 94 tokens per session scan A 07cacce98466

Subscribe to this mod's changes

api-security-tester is an agent published in the GitHub repository agigante80/actual-mcp-server (50 stars, last pushed yesterday), licensed MIT. It adds 94 tokens to every session and 2,520 once invoked, about $0.0005 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 2026-08-30.

Related

Other agents, from other repositories

tdd-guide

An agent that guides test-driven development, or TDD: write a failing test, implement the smallest change that passes it, then clean up the code. It covers unit, integration, and end-to-end tests and examines edge cases.

sangrokjung/claude-forge · 110 tokens

skeptical-auditor

Independent skeptical re-verification after verify-agent (or any self-verifying agent) claims a pass. Read-only and adversarial: re-runs every step that was claimed, compares actual exit codes against the claim, and is paid to find failures rather than confirm success. Never approves without executed evidence. Spawned…

sangrokjung/claude-forge · 120 tokens

e2e-runner

Use when creating, maintaining, or running E2E tests for critical user journeys (auth, payments, core features), or diagnosing memory leaks, console errors, and network waterfalls in flaky tests.

sangrokjung/claude-forge · 45 tokens

verify-agent

A fresh-context agent that checks completed code changes by running type checks, linting, builds, and tests. Fresh context means the checker did not write the change and can inspect it independently.

sangrokjung/claude-forge · 134 tokens

cos-compliance

Use this agent before shipping, merging, or deploying changes. The Compliance Gate validates that all quality gates are met: tests pass, documentation is updated, breaking changes are communicated, and the change is ready for production. Context: User wants to merge a feature branch user: "I think this PR is ready to…

winstonkoh87/Athena-Public · 180 tokens

verification-gate

Evidence-before-claims gate. Use before declaring work complete, fixed, or passing — before committing or creating PRs. Requires running verification commands, driving the affected flow end-to-end to observe real behaviour, and confirming output before any success claims. Adapted from Superpowers'…

sliamh11/Deus · 0 tokens