Frida (iOS) — wrapper

This is a wrapper around the upstream Anthropic skill. The full procedural reference is at:

• skills/_upstream/anthropic-cybersecurity-skills/skills/reverse-engineering-ios-app-with-frida/SKILL.md
• skills/_upstream/anthropic-cybersecurity-skills/skills/reverse-engineering-ios-app-with-frida/references/workflows.md
• skills/_upstream/anthropic-cybersecurity-skills/skills/reverse-engineering-ios-app-with-frida/scripts/agent.py

Read those for the canonical workflow.

How this repo extends the upstream skill

Upstream concept	Local extension
One-shot agent.py Frida orchestrator	frida_layer/runner.py — async, message-pump into JSONL store
Ad-hoc hook examples	frida_layer/hooks/*.js — pinned, versioned, parameterizable
JSON dumps to frida_ios/	runs/<run_id>/frida_events.jsonl + SQLite indexes for query
Per-script outputs	Correlated with HTTP flows via agent/correlate.py
Manual reporting	Auto-rendered Markdown findings via templates/finding.md.j2

Default hooks always loaded in our pipeline

• ssl_pinning_bypass.js
• jailbreak_bypass.js
• url_session_tracer.js
• commoncrypto_tracer.js

See frida_layer/auto_hook.py for the decision logic that adds more.

When to use this skill vs writing a one-off

• Use this skill for a sustained engagement where you want findings, correlation, and reports.
• Use the upstream skill standalone when you just want to grab a key from a binary and don't need the rest of the pipeline. The upstream agent.py is a great quick-start.