agentleFS
TheQmaksTheQmaks/areclaw/.claude/skills/intercept/SKILL.md

intercept

Skill67 starsChanged 9 months agoMITView on GitHub

Intelligent traffic interception — SSL bypass, HTTP logging, traffic collection

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No config file → default (user CA NOT trusted on Android 7+)

    line 27
  2. Very specific:

    Use phantom-frida stealth server.

    line 71
  3. Distinctive:

    Write a custom Frida script to hook the specific pinning implementation found in recon.

    line 92
  4. Very specific:

    If using http-logger.js, output goes to console.

    line 128

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/intercept — Intelligent Traffic Interception

You are setting up intelligent traffic interception. Not just "run mitmproxy" — choose the right strategy based on the app's protections.

Input

  • $ARGUMENTS — package name (must be installed on device)

Phase 1: Reconnaissance

1a. Check network security config

# Decompile resources
java -jar tools/apktool/apktool.jar d -s -f -o /tmp/intercept_recon $APK

Read res/xml/network_security_config.xml:

  • <domain-config cleartextTrafficPermitted="false"> → HTTPS only
  • <pin-set> → Certificate pinning configured
  • <trust-anchors><certificates src="user"/> → User CA trusted (easy mode!)
  • No config file → default (user CA NOT trusted on Android 7+)

1b. Detect pinning implementation

grep -rn 'CertificatePinner\|TrustManager\|X509\|ssl_pinning\|NetworkSecurityConfig' workspace/output/$PKG/

Determine pinning type:

  • None → Simple proxy is enough
  • OkHttp CertificatePinner → ssl-bypass.js handles it
  • Custom X509TrustManager → ssl-bypass.js handles it
  • Conscrypt/platform → ssl-bypass.js handles it
  • Native TLS (BoringSSL in .so) → Need Interceptor hook on SSL_CTX_set_verify
  • Flutter → Need special Flutter SSL bypass
  • React Native → Patch network_security_config + ssl-bypass.js

Phase 2: Strategy Selection

Based on recon, choose approach:

Strategy A: No Pinning

# Just log with Frida (simplest, no proxy needed)
frida -U -f $PKG -l workspace/frida-scripts/http-logger.js

Strategy B: Standard Pinning (OkHttp/TrustManager/Conscrypt)

# SSL bypass + HTTP logging
frida -U -f $PKG \
  -l workspace/frida-scripts/ssl-bypass.js \
  -l workspace/frida-scripts/http-logger.js

Strategy C: Heavy Pinning + Root Detection

# Full bypass stack
frida -U -f $PKG \
  -l workspace/frida-scripts/root-bypass.js \
  -l workspace/frida-scripts/ssl-bypass.js \
  -l workspace/frida-scripts/http-logger.js

Strategy C2: App Detects Frida (crash/illegal instruction on attach)

Use phantom-frida stealth server. Read tools/phantom-frida/build-info.json for name and port:

# Parse config
cat tools/phantom-frida/build-info.json
# → {"name": "<NAME>", "port": <PORT>, ...}

# Deploy phantom-frida to device (if not already running)
adb push tools/phantom-frida/<NAME>-server /data/local/tmp/<NAME>-server
adb shell chmod 755 /data/local/tmp/<NAME>-server
adb shell /data/local/tmp/<NAME>-server -D &
adb forward tcp:<PORT> tcp:<PORT>

# Use -H instead of -U (custom port)
frida -H 127.0.0.1:<PORT> -f $PKG \
  -l workspace/frida-scripts/ssl-bypass.js \
  -l workspace/frida-scripts/http-logger.js

Signs that app detects Frida: crashes immediately on frida -U, "illegal instruction" error, app shows "security check failed", or app force-closes when Frida attaches.

Strategy D: Native/Custom Pinning

Write a custom Frida script to hook the specific pinning implementation found in recon. May need to hook native functions in specific .so files.

Strategy E: mitmproxy (for full request/response bodies)

# Start mitmproxy with HAR dump
mitmdump -w workspace/traffic/$PKG.har --set stream_large_bodies=10m

# Configure device proxy
adb shell settings put global http_proxy <computer_ip>:8080

# Then bypass pinning
frida -U -f $PKG -l workspace/frida-scripts/ssl-bypass.js

Phase 3: Traffic Collection

3a. Start interception

Launch chosen strategy from Phase 2.

3b. Navigate the app

# Automated navigation
python pytools/ui_explorer.py snapshot  # See current state

# Walk key flows:
# - App startup (splash → main screen)
# - Login/registration
# - Main features
# - Settings
# - Profile

Spend adequate time on each screen. Some API calls happen on timers or scroll events.

3c. Save traffic

If using http-logger.js, output goes to console. Capture it:

frida -U -f $PKG -l workspace/frida-scripts/ssl-bypass.js -l workspace/frida-scripts/http-logger.js 2>&1 | tee workspace/traffic/$PKG-raw.log

# Extract HTTP-LOG lines
grep '^\[HTTP-LOG\]' workspace/traffic/$PKG-raw.log | sed 's/^\[HTTP-LOG\] //' > workspace/traffic/$PKG-traffic.json

Phase 4: Results

4a. Generate Postman collection

python pytools/traffic_to_collection.py frida workspace/traffic/$PKG-traffic.json \
  --name "$PKG API" \
  --output workspace/collections/$PKG.json

4b. Generate OpenAPI spec (if mitmproxy was used)

# If traffic was captured via mitmproxy (not Frida http-logger)
mitmproxy2swagger -i workspace/traffic/$PKG.flow -o workspace/collections/$PKG-api.yaml \
  -p https://api.example.com --examples

4c. Summary

Report to user:

  • Number of unique endpoints discovered
  • Base URLs found
  • Auth mechanism detected
  • Notable findings (interesting headers, tokens, unexpected calls)
  • Path to traffic file and Postman collection

Phase 5: Cleanup

# Remove proxy if set
adb shell settings put global http_proxy :0

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.