원클릭으로
cc-trace
Intercept and debug Claude Code API requests using mitmproxy. Inspect system prompts, tool definitions, token usage, and streaming responses. Triggers: cc-trace, mitmproxy, API interception, Claude Code debugging, traffic inspection.
메뉴
Intercept and debug Claude Code API requests using mitmproxy. Inspect system prompts, tool definitions, token usage, and streaming responses. Triggers: cc-trace, mitmproxy, API interception, Claude Code debugging, traffic inspection.
SOC 직업 분류 기준
Query and explore the 2600: The Hacker Quarterly magazine archive (1984-present) via DuckDB. Provides structured access to 168+ issues covering hacker culture, security, privacy, telephony, and digital rights without loading full content into context.
ACSets (Attributed C-Sets): Algebraic databases with Specter-style bidirectional navigation. Category-theoretic formalism for relational databases.
Attributed C-Sets as algebraic databases. Category-theoretic data structures generalizing graphs and dataframes with Gay.jl color integration.
ACSets (Attributed C-Sets): Algebraic databases with Specter-style bidirectional
Bridge active inference theory with robot control using K-Scale's JAX/MuJoCo stack. Use when connecting predictive coding to locomotion policies, mapping KL divergence minimization to RL training, applying mean field approximation to robotics state estimation, or implementing sim2real as inference about future observations.
Implement affective valence as directional derivative of interoceptive energy landscapes for AI alignment. Use when building alignment-aware RL agents, validating GF(3) conservation in reward signals, training Langevin-based policies, or analyzing fold-change detection signals in POMDP environments.
| name | cc-trace |
| description | Intercept and debug Claude Code API requests using mitmproxy. Inspect system prompts, tool definitions, token usage, and streaming responses. Triggers: cc-trace, mitmproxy, API interception, Claude Code debugging, traffic inspection. |
Capture and analyze Claude Code's API traffic using mitmproxy.
# Start mitmproxy filtering Claude API traffic
mitmweb --web-port 8081 --set flow_filter='~d api.anthropic.com' --save-stream-file ~/claude-flows.mitm
# In another terminal, launch Claude through the proxy
proxy_claude
Add to ~/.zshrc or ~/.bashrc:
proxy_claude() {
export HTTP_PROXY=http://127.0.0.1:8080
export HTTPS_PROXY=http://127.0.0.1:8080
export NODE_EXTRA_CA_CERTS="$HOME/.mitmproxy/mitmproxy-ca-cert.pem"
export NODE_TLS_REJECT_UNAUTHORIZED=0
claude
}
# Generate cert (start mitmproxy once, then quit)
mitmproxy --set flow_filter='~d api.anthropic.com'
# Trust cert in system keychain
sudo security add-trusted-cert -d -p ssl -p basic -k /Library/Keychains/System.keychain ~/.mitmproxy/mitmproxy-ca-cert.pem
| Filter | Description |
|---|---|
~d api.anthropic.com | Only Anthropic API |
~m POST | Only POST requests |
~s "tool_use" | Responses containing tool_use |
~b "system" | Body contains "system" |
| Combine | ~d api.anthropic.com & ~m POST & ~b "tool_use" |
# Dump all captured requests
mitmdump -r ~/claude-flows.mitm -n --set flow_detail=2
# Show last prompt sent
bash scripts/show-last-prompt.sh
from mitmproxy import http
import json
def response(flow: http.HTTPFlow):
if "api.anthropic.com" in flow.request.pretty_host:
req = json.loads(flow.request.get_text())
print(f"Model: {req.get('model')}, Messages: {len(req.get('messages', []))}, Tools: {len(req.get('tools', []))}")
| Script | Purpose |
|---|---|
scripts/verify-setup.sh | Check mitmproxy install, cert trust, shell config |
scripts/parse-streamed-response.ts | Parse Anthropic SSE format |
scripts/extract-slash-commands.py | Extract user messages from flows |
scripts/show-last-prompt.sh | Show most recent user prompt |
security find-certificate -c mitmproxy -a # verify cert
lsof -i :8080 # check port
curl -x http://127.0.0.1:8080 https://api.anthropic.com 2>&1 | head -5