Generate and query a static security call graph from source code.
npx skills add https://github.com/ayehiaa/my-travel-assistant --skill trailmark
Please help me install this Agent Skill. Skill: trailmark Source: https://github.com/ayehiaa/my-travel-assistant/tree/main/.agents/skills/trailmark Command: npx skills add https://github.com/ayehiaa/my-travel-assistant --skill trailmark
Trailmark turns source code into a queryable security graph so you can trace how inputs flow through functions to find attack surface, taint propagation, privilege boundaries, and blast radius before deeper review.
Use the trailmark skill to build a security call graph from a target directory by running the command: uv run trailmark analyze --language auto --summary {targetDir}
High-intent search queries and answers about installing and using this skill.
To trace user input reachability to sensitive functions, generate a static security call graph to map attack surface and taint propagation across your codebase. This models how data flows through functions to identify blast radius before deeper review.
Yes, you can perform taint propagation analysis across a polyglot codebase by building a multi-language static security call graph. It supports explicit or auto language selection and models edge confidence for defensible security claims.
The best way to prioritize security findings in an unfamiliar repository is running pre-analysis subgraphs to estimate blast radius, enumerate entry points, and detect privilege boundaries. This highlights the highest-risk nodes and call paths for focused review.
Generate a security call graph by running the analysis command with auto language selection and summary output on the target directory. This populates blast radius, privilege boundary, and tainted subgraphs for downstream security queries.
Static code graph analysis works for identifying privilege boundaries by parsing functions, classes, and modules into a queryable graph. It maps call paths and applies pre-analysis to detect where sensitive boundaries exist across the codebase.
Edge confidence modeling improves attack surface analysis by distinguishing certain, inferred, and uncertain call relationships in the security graph. This differentiation ensures more defensible security claims when tracing taint propagation and blast radius.
InverterNetworkAnalyze repositories with pygount to calculate lines of code and language distribution.
SuperlendParse SARIF 2.1.0 files and extract structured findings for reporting.
iiammaeConstruct and query source code graphs for security analysis.
AcczdyTest iOS and Android apps for vulnerabilities using OWASP MASTG.
