agentsclimarketplace

Exa security basics

Skill jeremylongshore/claude-code-plugins-plus-skills/plugins/saas-packs/exa-pack/skills/exa-security-basics

425 plugins, 2,810 skills, 200 agents for Claude Code. Open-source marketplace at tonsofskills.com with the ccpi CLI package manager.

Install
npx -y skills add jeremylongshore/claude-code-plugins-plus-skills --skill exa-security-basics

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

'Secure Exa API keys, implement content moderation, and manage domain restrictions.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

5.5 KB, as published. Nobody here has run it

Exa Security Basics

Overview

Security best practices for Exa API integrations. Exa authenticates via the x-api-key header. Key security concerns include API key protection, content moderation for search results, domain filtering to prevent exposure to malicious sources, and query sanitization.

Prerequisites

  • Exa API key from dashboard.exa.ai
  • Understanding of environment variable management
  • .gitignore configured for secrets

Instructions

Step 1: API Key Management

# .env (NEVER commit to git)
EXA_API_KEY=your-api-key-here

# .gitignore — add these entries
.env
.env.local
.env.*.local
// Validate API key exists before creating client
import Exa from "exa-js";

function createSecureClient(): Exa {
  const apiKey = process.env.EXA_API_KEY;
  if (!apiKey) {
    throw new Error("EXA_API_KEY not configured");
  }
  if (apiKey.startsWith("sk_") && apiKey.length < 20) {
    throw new Error("EXA_API_KEY appears malformed");
  }
  return new Exa(apiKey);
}

Step 2: Enable Content Moderation

const exa = new Exa(process.env.EXA_API_KEY);

// Exa supports content moderation to filter unsafe results
const results = await exa.searchAndContents(
  "user-provided search query",
  {
    numResults: 10,
    text: true,
    moderation: true,  // filter unsafe content from results
  }
);

Step 3: Domain Filtering for Safety

// Restrict results to trusted domains for sensitive use cases
const TRUSTED_DOMAINS = [
  "docs.python.org", "developer.mozilla.org", "nodejs.org",
  "github.com", "stackoverflow.com", "arxiv.org",
];

const BLOCKED_DOMAINS = [
  "known-malware-site.com", "phishing-domain.net",
];

async function safeDomainSearch(query: string) {
  return exa.searchAndContents(query, {
    numResults: 10,
    includeDomains: TRUSTED_DOMAINS,  // only return results from these
    text: { maxCharacters: 1000 },
  });
}

async function searchWithBlocklist(query: string) {
  return exa.searchAndContents(query, {
    numResults: 10,
    excludeDomains: BLOCKED_DOMAINS,  // never return results from these
    text: { maxCharacters: 1000 },
  });
}

Step 4: Query Sanitization

// Sanitize user-provided queries before sending to Exa
function sanitizeQuery(input: string): string {
  // Remove potential injection patterns
  let clean = input
    .replace(/[<>{}]/g, "")           // strip HTML/template chars
    .replace(/\0/g, "")              // remove null bytes
    .trim()
    .substring(0, 500);              // cap query length

  if (!clean || clean.length < 2) {
    throw new Error("Query too short or empty after sanitization");
  }
  return clean;
}

// Usage
const userQuery = sanitizeQuery(req.body.query);
const results = await exa.search(userQuery, {
  numResults: 10,
  moderation: true,
});

Step 5: Per-Environment Key Isolation

// Use separate API keys per environment
const KEY_MAP: Record<string, string> = {
  development: process.env.EXA_API_KEY_DEV!,
  staging: process.env.EXA_API_KEY_STAGING!,
  production: process.env.EXA_API_KEY_PROD!,
};

function getExaForEnv(): Exa {
  const env = process.env.NODE_ENV || "development";
  const key = KEY_MAP[env];
  if (!key) throw new Error(`No EXA key for ${env}`);
  return new Exa(key);
}

Security Checklist

  • API key stored in environment variables (never hardcoded)
  • .env files in .gitignore
  • Separate API keys for dev/staging/production
  • moderation: true enabled for user-facing search
  • Query input sanitized before API calls
  • Domain allowlist/blocklist applied for sensitive use cases
  • API key rotation procedure documented
  • Git history scanned for accidentally committed keys

Error Handling

Security IssueDetectionMitigation
Exposed API keygit log -p searchRotate key immediately at dashboard.exa.ai
Unsafe search resultsUser reportsEnable moderation: true
Untrusted domainsReview result URLsApply includeDomains filter
Query injectionInput validationSanitize before search

Examples

Scan Git History for Leaked Keys

set -euo pipefail
# Check if API key was ever committed
git log -p --all -S "EXA_API_KEY" -- "*.ts" "*.js" "*.py" "*.env" | head -20

Key Rotation Procedure

set -euo pipefail
# 1. Generate new key in dashboard.exa.ai
# 2. Update environment
export EXA_API_KEY="new-key-here"
# 3. Verify new key works
curl -s -o /dev/null -w "%{http_code}" \
  -X POST https://api.exa.ai/search \
  -H "x-api-key: $EXA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query":"test","numResults":1}'
# 4. Revoke old key in dashboard

Resources

Next Steps

For production deployment, see exa-prod-checklist.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.