agentleFS
thejefflarsonthejefflarson/soundcheck/.claude/skills/redos/SKILL.md

redos

Skill20 starsChanged 4 months agoMITView on GitHub

Detects regex patterns vulnerable to catastrophic backtracking on crafted

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Bound regex execution: When the engine supports it, set a match timeout or run the match on a cancellable worker.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

ReDoS Security Check (CWE-1333)

What this checks

Protects against Regular Expression Denial of Service where a crafted input causes catastrophic backtracking in regex engines. A single malicious string can pin a CPU core for minutes or hours, taking down the service.

Vulnerable patterns

  • Nested quantifiers applied to user input — a quantified group whose body is itself quantified produces exponential state
  • Overlapping alternation inside repetition — alternatives that can match the same prefix multiply backtracking work
  • Common "email" or "URL" regexes built from quantified character classes wrapped in further repetition
  • Regexes compiled from external configuration or user input where the pattern shape itself is attacker-controlled

Fix immediately

Flag the vulnerable regex, explain the risk, and suggest a fix establishing these properties. Translate the principle to the language and engine of the audited file — do not echo examples from another stack.

  1. No nested quantifiers on user input. Collapse a quantified group whose body is also quantified into a single quantifier. Rewrite alternations so alternatives cannot match the same prefix.
  2. Prefer a linear-time regex engine when one is available for the language under review. Engines that disallow backtracking guarantee linear matching regardless of pattern shape.
  3. Bound regex execution. When the engine supports it, set a match timeout or run the match on a cancellable worker; when it does not, simplify the pattern until worst-case input completes in bounded time.
  4. Treat patterns sourced from configuration or user input as untrusted. Either pre-compile a fixed allowlist of patterns, or run user-supplied patterns on a sandbox engine with a hard timeout.

Verification

  • No regex applied to user input contains nested quantifiers or alternation inside repetition where alternatives share a prefix
  • If the language offers a linear-time engine, the audited code uses it for user-input regexes; otherwise patterns are simplified or execution is time-bounded
  • Regex patterns loaded from external configuration or user input run against an allowlist or under an enforced timeout

References

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.