Skip to main content
skill CX Codex

taint-trace — Claude Code skill: taint analysis and installation

The taint-trace Claude Code skill adjusts a project's static taint-analysis models, then runs its analyzer. It addresses missed data paths, unresolved framework calls, and false positives caused by imprecise models.

CC-BY-SA-4.0 130 installs ★ 8,341 updated 2026-10-24

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add trailofbits/skills --skill taint-trace -a claude-code
Open on GitHub

What taint-trace does

taint-trace draws on Trailmark graph data to revise how the analyzer represents input sources, risky operations, sanitization, value transfer, and types. It then runs the project's existing analyzer. The skill stores the initial and updated run results, model-change records, and a readable report in .trailmark/taint-trace/. The initial run is retained.

Its helper runs through uv and gets trailmark>=0.8.0,<0.9 from a package index; the native analyzer is a separate dependency. The selected engine's capabilities and models are checked before analysis. Analyzer startup can execute project code through autoloaders, plugins, or configuration hooks, and static analysis does not isolate that code.

What it helps with

  • Trace external input to a risky operation
  • Restore data paths hidden by incomplete models
  • Resolve framework calls the analyzer cannot bind
  • Reduce false positives caused by imprecise models

Example prompts for the agent

  • Analyze data flows in src and run /trailmark:taint-trace-run.
  • Find missed paths from external input to SQL queries in app.
  • Investigate false positives in backend and save a report.

How to install taint-trace

1Skills CLI (Vercel)

npx skills add trailofbits/skills --skill taint-trace -a claude-code

2Manually: put the files in place

~/.claude/skills/taint-trace/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/taint-trace/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about taint-trace

How do I install taint-trace in Claude Code?

Run npx skills add trailofbits/skills --skill taint-trace -a claude-code. The skill file is located at ~/.claude/skills/taint-trace/SKILL.md.

What does taint-trace do?

It uses Trailmark graph data to revise models for a project's static taint analyzer, then runs that analyzer. Results, model changes, and a report are saved under .trailmark/taint-trace/.

What does taint-trace need to run?

It uses uv to install trailmark>=0.8.0,<0.9 from a package index, and requires a separate native analyzer. The selected engine's capabilities are checked before analysis.

Similar skills

All skills for Security →

From the same set95

The whole set →
prove-patch prove-patch — a skill for Claude Code validates a local security patch with scoped cases and a three-arm comparison. It runs when a patch commit needs independent proof. CX Codex +1 676installs hunt-regressions hunt-regressions — a skill for Claude Code that discovers potential regressions before reproduction. Runs when analyzing code changes and their callers. CX Codex +1 675installs reproduce-regression The reproduce-regression Claude Code skill reproduces a regression candidate on live systems before and after a change. It builds a PoC and compares both versions using the same input. CX Codex +1 671installs report-regressions report-regressions — a skill for Claude Code builds regression reports from saved campaign decisions and links findings to evidence. It runs when a campaign report is created or refreshed. CX Codex +1 670installs prepare-regression prepare-regression — a skill for Claude Code that pins source revisions and gathers commit or PR context for a regression investigation. It runs during investigation setup. CX Codex +1 670installs adjudicate-regression adjudicate-regression — a skill for Claude Code and Codex that reviews reproduction evidence to confirm whether a flaw is a true regression or an exclusion. It evaluates live artifacts before logging issues. CX Codex +1 669installs replay-regression The replay-regression Claude Code skill independently tests a supplied PoC against fresh builds from before and after a change. It runs the same input on S0 and S1, then records what happened. CX Codex +1 668installs cvss-scoring The cvss-scoring Claude Code skill turns a vulnerability description into a CVSS v4.0 vector. It activates for requests to score a finding or estimate its severity. CX Codex +1 304installs

Project sponsors

Companies that keep this analytics open