Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    API Penetration Testing

    1

    Authorized API pentest that drives toward a real objective, not a checklist

    $24

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    API Penetration Testing

    API Penetration Testing

    Example session with this skill installed

    Test for IDOR/BOLA on my endpoints

    • Read your context and instructions
    • Compiled the api penetration testing

    EXECUTION CLASSIFICATION

    • USER-SUPPLIED FACTS: test objective (IDOR/BOLA).
    • OBSERVED EVIDENCE: none.
    • HYPOTHETICAL / EXAMPLES: candidate REST API endpoints below are marked [HYPOTHETICAL EXAMPLE].
    • UNVERIFIED / MISSING INPUTS: authorization statement, target service URL, environment, test identities/tokens, scope limits.

    EXECUTION MODE: NONE — BLOCKED

    ENGAGEMENT

    • AUTHORIZATION: MISSING / NOT SUPPLIED (Active testing blocked)
    • TARGETS: NOT SUPPLIED
    • ENVIRONMENT: UNVERIFIED / NOT SUPPLIED
    • SCOPE: NOT SUPPLIED

    EVIDENCE STATUS: BLOCKED — MISSING AUTHORIZATION/SCOPE
    CONFIRMED HTTP FINDING: NO
    OBSERVED: NONE — NO ACTIVE RUNTIME EXECUTION PERFORMED

    ANALYSIS & HYPOTHETICAL TEST PLAN:
    Active testing cannot proceed without explicit authorization, target URL, and scoping boundaries.

    Below is an illustrative test structure for when authorization and scope are provided:

    [HYPOTHETICAL EXAMPLE TEST CASE]

    1. Target: [HYPOTHETICAL] https://target-api.local/api/v1/resources/{id}
    2. Identities needed: [HYPOTHETICAL] User A (owner), User B (non-owner)
    3. Step 1 (Planned Baseline): User A requests owned resource -> expected 200
    4. Step 2 (Planned BOLA probe): User B requests User A's resource -> expected 403/404

    LIMITATIONS: No runtime commands executed. No responses observed. No fabricated status codes generated.

    NEXT STEPS: Provide written authorization, target endpoint/environment, and valid test accounts.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    What it does

    • Authorized testing of API surfaces including REST, GraphQL, RPC, and webhooks.
    • Authorization boundary checks to verify IDOR, BOLA, and cross-tenant data leaks.
    • Business logic analysis to identify unsafe state transitions and chained API abuse.
    • Token validation for scope confusion and broken function-level authorization.
    • Evidence-backed reporting that distinguishes between observed facts and hypotheses.

    How it works

    1. Scope verification to confirm explicit authorization, target URLs, and test identities.
    2. Baseline capture of authorized requests through a real HTTP adapter or local harness.
    3. Variable manipulation by changing security-relevant parameters one at a time.
    4. Impact proofing via observed request/response evidence with PII redaction.
    5. Finding classification to output a verdict based on reproducible server-side enforcement.

    Frameworks & tools

    This skill is designed for REST, GraphQL, and RPC architectures. It utilizes real HTTP adapters for runtime evidence and integrates with existing project security requirements or Rules of Engagement.

    Why this beats prompting it yourself

    Standard LLM prompts often hallucinate security vulnerabilities or ignore strict scoping boundaries. This skill enforces a mandatory execution contract and evidence discipline, ensuring that every finding is backed by observable data and that testing stops immediately upon reaching the defined objective.

    Validation

    Routing verified in isolated Antigravity sessions — 10/10 target-specific blind-pilot cases passed, plus 2/2 post-hardening evidence-discipline regressions with zero unsupported factual claims.

    Use cases

    • Test if one tenant can access another's resources via IDOR/BOLA.
    • Validate that GraphQL schemas do not expose excessive data to guest users.
    • Verify that administrative functions are protected against token scope confusion.
    • Check webhook signature verification and replay protection.

    Known limitations

    Does not perform browser UI testing, code-only static review (SAST), or destructive operations like DDoS. Requires explicit user-supplied authorization before any active execution begins.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.

    Active testing requires explicit user supplied authorization, a defined target, and sufficient execution context. The skill does not fabricate runtime evidence when a live adapter or authorized test environment is unavailable. It is focused on API surfaces and does not replace browser focused web testing, static code review, or broader infrastructure assessment.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 25 days ago

    Permissions

    Read Files
    Network Access
    Terminal / Shell

    Project access is task-scoped; no fixed file path or external account is required. Do not grant broader authority than the current task. Active testing requires explicit user-supplied authorization and scope. Credentials/tokens must be supplied for the authorized target; the skill does not harvest environment variables.

    Listed25 days ago
    Updated23 days ago

    What's inside

    Frequently Asked Questions

    Runs an objective-driven penetration test across REST, GraphQL, RPC and webhook surfaces — object/function authorization, token boundaries, schema and business-logic abuse, chained to prove a defined objective. Ships with a mandatory inline authorization-and-scope gate and a bounded reporting contract, so it produces a scoped finding with evidence, never an open-ended scan. Includes a worked example end to end.

    Use it for: Pentest a REST API for IDOR/BOLA; Check GraphQL authorization boundaries; Test cross-tenant API access.

    Not for browser UI testing or code-only review.

    It may perform bounded active testing only after explicit authorization and scope are supplied. It does not modify project artifacts, commit, push, merge, deploy, or perform destructive actions.

    Works with any SKILL.md-compatible agent. Active execution requires explicit authorization, scoped targets, and an agent with the needed HTTP/network capability. Without runtime evidence, results remain BLOCKED/UNVERIFIED.

    New in Security