🔍 API Pulse — The Developer's API Swiss Army Knife

Probe, benchmark, compare, and diagnose any API — from your terminal.

One tool. Four modes. Zero fluff.

$ python api_pulse.py probe https://api.stripe.com
$ python api_pulse.py bench https://your-api.com/endpoint -n 200 -c 20
$ python api_pulse.py compare https://api.github.com https://api.gitlab.com
$ python api_pulse.py chain workflow.json

Why this exists

Postman is bloated. curl gives you raw bytes. wrk only does load testing. httpie only does pretty requests. None of them do all of this in one tool:

Feature	curl	httpie	wrk/hey	Postman	API Pulse
Single request with details	✅	✅	❌	✅	✅
TLS/SSL certificate inspection	❌	❌	❌	❌	✅
Security header audit + score	❌	❌	❌	❌	✅
Concurrent benchmarking	❌	❌	✅	❌	✅
Latency percentiles (p95/p99)	❌	❌	✅	❌	✅
Side-by-side API comparison	❌	❌	❌	❌	✅
Multi-step chain workflows	❌	❌	❌	✅	✅
Variable extraction between steps	❌	❌	❌	✅	✅
Beautiful terminal output	❌	✅	❌	N/A	✅
Standalone HTML reports	❌	❌	❌	❌	✅
Latency grading (A+ to F)	❌	❌	❌	❌	✅
Zero config / instant start	✅	✅	✅	❌	✅

The Four Modes

1. 🔍 Probe — Deep-inspect any endpoint

python api_pulse.py probe https://api.github.com

One request. Full X-ray. You get:

• Response status, latency, size
• Important response headers
• TLS certificate details (issuer, cipher, days until expiry)
• Security header audit with a 0-100 score
• Missing security headers flagged by severity (high/medium/low)

Add --html report.html to generate a shareable visual report.

2. ⚡ Bench — Stress-test with stats

python api_pulse.py bench https://your-api.com/endpoint -n 200 -c 20

Fire 200 requests with 20 concurrent workers. You get:

• Average, median, p95, p99, min, max latency
• Standard deviation (consistency)
• Requests per second
• Status code distribution with visual bars
• ASCII latency histogram
• A letter grade (A+ to F)

3. 🔀 Compare — Race endpoints head-to-head

python api_pulse.py compare https://api.github.com https://api.gitlab.com https://api.bitbucket.org

Benchmarks each endpoint with the same parameters and shows a side-by-side comparison table with the winner marked for each metric. Great for:

• Choosing between API providers
• Comparing your staging vs production
• Testing regional endpoints (US vs EU vs APAC)

4. 🔗 Chain — Multi-step API workflows

python api_pulse.py chain examples/chain_workflow.json -v

Run a sequence of API calls where each step can extract values and pass them to the next. Define workflows in simple JSON:

[
  {
    "name": "Create user",
    "url": "https://api.example.com/users",
    "method": "POST",
    "body": "{\"name\": \"Jane\"}",
    "extract": { "user_id": "$.id" }
  },
  {
    "name": "Get created user",
    "url": "https://api.example.com/users/{{user_id}}",
    "method": "GET"
  }
]

Use cases: testing auth flows, signup → verify → login chains, CRUD lifecycle tests, webhook simulation.

Install

git clone https://github.com/adithyakiranhere/api-pulse.git
cd api-pulse
pip install -r requirements.txt

Only dependency is rich for terminal output. Core HTTP engine uses Python's stdlib (urllib) — no requests needed.

HTML Reports

Add --html report.html to probe or bench commands to generate a gorgeous standalone HTML report. No server needed — just open the file in any browser.

python api_pulse.py probe https://api.stripe.com --html stripe-report.html
python api_pulse.py bench https://your-api.com -n 100 --html bench-report.html

Security Header Audit

The probe mode checks for 7 critical security headers and scores your API:

Header	Severity	What it prevents
Strict-Transport-Security (HSTS)	🔴 High	Downgrade attacks
Content-Security-Policy (CSP)	🔴 High	XSS and injection
X-Content-Type-Options	🟡 Medium	MIME sniffing
X-Frame-Options	🟡 Medium	Clickjacking
Permissions-Policy	🟡 Medium	Browser feature abuse
X-XSS-Protection	⚪ Low	Legacy XSS (use CSP instead)
Referrer-Policy	⚪ Low	Referrer leakage

Score weights: High = 25pts, Medium = 15pts, Low = 10pts. Perfect = 100/100.

Latency Grades

Grade	Avg Latency	Verdict
A+	< 100ms	Blazing fast
A	< 200ms	Excellent
B	< 500ms	Good
C	< 1000ms	Needs work
D	< 3000ms	Slow
F	> 3000ms	Critical

Use Cases

• Before deploying: probe your API to catch missing security headers
• During development: benchmark to find performance regressions
• Choosing providers: compare competing API services head-to-head
• Testing workflows: chain multi-step flows (auth → CRUD → cleanup)
• Incident response: quick probe to check if an API is responding and how fast
• CI/CD integration: run benchmarks in pipelines, fail on grade drops

Ideas for contributions

• Export benchmark results to JSON/CSV for data analysis
• WebSocket support for real-time API testing
• GraphQL query support
• Authentication helpers (OAuth2, API keys, JWT)
• CI mode with exit codes based on latency thresholds
• Watch mode — continuously probe and alert on degradation
• Compare results over time (save + diff benchmarks)
• Request body templating with Faker-generated data

License

MIT

---

Built because every developer deserves better than curl -v | grep and a prayer. 🔍