firish/webfetch

Python·136 files·commit 3399799·scanned 4d ago·cached ✓

70/100

DANGEROUS

Sensitive capability combinations reachable from entry points. Do not run without review.

This repository contains patterns associated with malware. Do NOT run this code.

verdict accurate?

Crypto/Web3 project

Research / educational context

threat-state: dangerouslive

FINDINGS ░▒▓

criticaleval() usage detectedeval() executes arbitrary code and is commonly used in malware to run obfuscated payloads fetched from remote servers.evals/run_matcher_eval.py

"""
Layer 1: offline semantic-matcher eval (no network beyond model downloads).

Answers the question the semantic cache needs answered before it exists:

criticalPython dynamic code executioneval(), exec() or compile() on untrusted input can execute arbitrary code. Common in Python malware for running obfuscated payloads.evals/run_matcher_eval.py

"""
Layer 1: offline semantic-matcher eval (no network beyond model downloads).

Answers the question the semantic cache needs answered before it exists:

criticaleval() usage detectedeval() executes arbitrary code and is commonly used in malware to run obfuscated payloads fetched from remote servers.evals/run_pipeline_eval.py

"""
Layer 2: live pipeline eval (DDG network, no LLM calls).

Runs diverse factoid queries with known short answers through

criticalPython dynamic code executioneval(), exec() or compile() on untrusted input can execute arbitrary code. Common in Python malware for running obfuscated payloads.evals/run_pipeline_eval.py

"""
Layer 2: live pipeline eval (DDG network, no LLM calls).

Runs diverse factoid queries with known short answers through

infoSuspicious files are not reachable from entry pointsFlagged files exist but are not imported by any entry point. They may be dead code, tests, or attack payloads triggered by another mechanism.+2

Scores are heuristics. A “safe” verdict means no known-malicious patterns were found — clever malware can look boring. Wrong verdict? Flag it above; confirmed false positives become regression tests.

107 files scanned @ 3399799 | 10/25/2026 | heuristic scan — always review manually

risk by category

code execution25

network & exfiltration0

file system access0

obfuscation0

supply chain0

owasp / injection0

telemetry

files 107/136rules hit 5engine v7commit 3399799

github

firish/webfetch

Own your LLM's web search: a local search->fetch->rank pipeline that replaces hosted web-search tools. Measured: matches hosted accuracy at 66% lower cost and up to 88% fewer tokens, plus a precision-tuned semantic caching with query-dependant TTL that no API offers.

Python

193d

136 files

107 scanned(79%)

3399799

architecture░▒▓

entry (0) flagged (2) pkg (87)

148 nodes · 355 edgesscroll to zoom · click node to jump to finding