tree-sitter-analyzer
Cross-language code indexing, call graphs, and safe refactoring
All Skills in This Repository (28)
Pure Emerald Level Indicatorsbenchmark
Measures web performance regressions by comparing navigation and resource timing data against baselines.
plan-design-review
Rate design dimensions from 0 to 10 and identify quality gaps in design plans.
autoplan
Load CEO, design, and engineering review skills to automate plan reviews.
design-consultation
Generate a complete design system with tokenized typography, color, spacing, motion, and DESIGN.md.
freeze
Deny Edit or Write operations outside a configured directory path.
careful
Intercept destructive shell and deployment commands for confirmation before execution.
cso
Assess security posture across code, infrastructure, and processes.
canary
Monitor production URLs for errors, performance regressions, and broken pages.
investigate
Identifies and analyzes root causes of software bugs through structured investigation.
document-release
Update project documentation files from branch diffs after PR merges.
gstack-upgrade
Upgrade gstack to the latest version and sync vendored copies.
land-and-deploy
Merge PRs after CI passes and trigger production deployment with canary health checks.
Frequently Asked Questions
FAQPage SchemaHow to install tree-sitter-analyzer?▼
Run `npx skills add aimasteracc/tree-sitter-analyzer --all -g -y` in your terminal to install all skills in this suite globally.
How to check if a code change is safe?▼
The tsa-edit-safety skill runs a pre-edit gate that returns a SAFE, REVIEW, CAUTION, or UNSAFE verdict plus the exact test command to run, replacing manual grep and git-diff exploration.
How to find who calls a function?▼
The tsa-graph skill answers who-calls-X, what-X-calls, and call-path questions in one query using a persisted cross-file call graph instead of slow grep searches.
Does tree-sitter-analyzer work with Claude Code and Cursor?▼
Yes. It runs as an MCP server and CLI that works with Claude Code, Cursor, and any MCP-compatible client, with token-efficient TOON output by default.
Can it review a pull request automatically?▼
Yes. The tsa-pr-review skill classifies each changed symbol, measures blast radius from the call graph, checks architecture rules, and returns an APPROVE, REVIEW, or BLOCK verdict with the exact pytest command to gate the merge.
Related Repositories in Software Engineering
View All in Software Engineering→openclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core