Bandit
6,000Security linter for Python. Catches common code-level vulnerabilities.
Maturity tier details
Computed tier: 2B(raw score 4.51). Effective tier: 2B.
Facets:documented_in=0evaluated_on=0integration_surfaces=1paper_citations=0recent_commit=2stars=1.51
Source of truth:pipeline/maturity/score.py
Description
Bandit is a Python security linter from PyCQA: it walks the AST looking for common insecure patterns — hardcoded passwords, use of `eval`/`exec`, weak crypto primitives, shell-injection-prone subprocess calls, insecure deserialization, and so on. It runs from the CLI and in CI as a post_generation security gate. Configuration lives in `pyproject.toml [tool.bandit]`, which the audit detector now recognises.
Why it's included
Occupies detection × post_generation × fully_autonomous for security_vulnerability in the Python lane. Distinct from Semgrep because Bandit is Python-specific with hand-tuned rules covering Python's standard library footguns; including it lets the audit detector resolve `[tool.bandit]` blocks.
Security notes
Standard dev-tool security posture; Bandit runs locally against source and does not execute target code. Like all rule-based scanners it produces both false positives and false negatives — pair it with a complementary scanner (e.g. semgrep) on regulated codepaths.
Addresses failure modes
- Security vulnerability — Insecure patterns, injection flaws, weak cryptography, secret leakage.