More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    API Security Assessment

    1

    Assess your own API the way an attacker would: four lanes covering discovery, authentication, authorization, injection, and abuse, with every finding filed with a fix and a regression test.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    API Security Assessment

    API Security Assessment

    Example session with this skill installed

    Run the quick version of the assessment on our staging API — a small REST API for a shop backend — and file whatever you find.

    • Read your context and instructions
    • Compiled the api security assessment

    SCOPE — Staging environment, two test accounts (same role), one low-privilege account, load ceiling agreed: no abuse tests beyond 20 concurrent requests.

    SMOKE PASS RESULTS

    1. Object replay (Lane 2) — FAIL. As account B, GET /v1/orders/4471 (account A's order) returned 200 with the full order. Five minutes in, the most common real-world finding is present.
    2. Over-returned fields (Lane 2) — FAIL. The order response includes customer_email and an internal cost_margin field the storefront UI never displays.
    3. Admin endpoint as low-privilege user (Lane 2) — PASS. GET /v1/admin/refunds returned 403.
    4. Token tampering (Lane 1) — PASS. Altered exp claim rejected, token lifetime 1 hour as configured.
    5. Pagination push (Lane 2) — FAIL. ?limit=100000 on /v1/orders returned the full list in one response, no cap.
    6. Not applicable — REST only, no GraphQL surface found.

    FINDINGS FILED (template format):

    • F-01 Critical, owner: API lead. GET /v1/orders/:id — broken object-level authorization lets any authenticated user read any customer's order, including email and pricing internals. Fix at the authorization layer: per-order ownership check on every read. Regression test: two-session replay added to CI.
    • F-02 High. Response over-exposure on the same endpoint (email + cost_margin). Fix: response allow-list.
    • F-03 High. Uncapped pagination on list endpoints. Fix: hard cap at 100, return 400 above it.

    DECISION — Two authorization failures in the smoke pass skip the queue: full Lane 2 walk across every object type before anything else, per the skill's decision rule. Lanes 1, 3, and 4 proceed after the Lane 2 inventory is complete; every fix gets a replay retest before its finding closes.

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

    About this skill

    A defensive assessment method for APIs you own or are explicitly authorized to test, organized the way an attacker actually moves. Lane 0 sets scope, written authorization, evidence discipline, and a testing load ceiling before anything is touched. Lane 1 maps the real attack surface — not just the documented spec, but undocumented versions, forgotten endpoints, and exposed schemas — then attacks the identity layer: token tampering, algorithm swaps, OAuth flow flaws, and WebSocket handshake gaps. Lane 2 is where the money findings live: two-session object-level authorization replays, function-level checks with low-privilege accounts, mass assignment, and response over-exposure, including legacy SOAP surfaces most audits skip. Lane 3 covers injection at every input, business-logic flaws like skipped state transitions and race conditions on scarce resources, and rate-limit abuse including GraphQL query-cost attacks. Lane 4 turns the one-time audit into a regression loop: authorization replays automated in CI, spec-change triggers, and severity-gated merges. Every finding is filed on a standard template with a one-sentence impact statement and rated on a severity decision table, so reports get fixed instead of debated. A 20-minute smoke pass finds the most common real findings fast, and a worked example traces four findings through discovery, rating, fix, and retest. A section on findings to not chase keeps scanner noise from eating the remediation budget.

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

    • It directs and structures testing; it does not itself execute requests against your API — you run the replays in your own tooling and bring back the traffic.
    • Severity ratings are calibrated judgment, not a compliance certification; regulated environments may also need a formal audit.
    • Business-logic coverage depends on how well you can describe your workflows and roles.

    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 24 days ago

    Permissions

    No special permissions declared or detected

    Listed24 days ago
    Updated21 days ago

    What's inside

    Frequently Asked Questions

    Assess your own API the way an attacker would: four lanes covering discovery, authentication, authorization, injection, and abuse, with every finding filed with a fix and a regression test.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use API Security Assessment on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in APIs & Backend