Borrowing it
Nothing to install: this file belongs to aimasteracc/tree-sitter-analyzer. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/aimasteracc/tree-sitter-analyzer/main/.claude/skills/tsa-pr-review/SKILL.mdgit clone --depth 1 https://github.com/aimasteracc/tree-sitter-analyzerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aimasteracc/tree-sitter-analyzer/tsa-pr-review)<a href="https://agentmods.dev/skills/aimasteracc/tree-sitter-analyzer/tsa-pr-review"><img src="https://agentmods.dev/badge/skills/aimasteracc/tree-sitter-analyzer/tsa-pr-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aimasteracc/tree-sitter-analyzer/tsa-pr-review"><img src="https://agentmods.dev/badge/skills/aimasteracc/tree-sitter-analyzer/tsa-pr-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00354 | $0.02980 |
| Opus 5 | $0.00177 | $0.01490 |
| Sonnet 5 | $0.00071 | $0.00596 |
| Haiku 4.5 | $0.00035 | $0.00298 |
Grade A, and why
tsa-pr-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 288 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tsa-pr-review — AST-grounded PR review with merge-gate verdict
Replaces "LLM reads the diff and guesses" with a deterministic pipeline that grounds the review in (a) the AST diff, (b) the persisted call graph, and (c) architectural-constraints.yml. Output is a structured verdict an agent can act on — including the exact pytest command to run before merge.
When to use
- Local diff review: unstaged, staged, or branch-vs-main
- GitHub PR review by URL (uses
ghunder the hood) - CI / agent pre-merge gate
- Any time a generic "read the diff" review would be too shallow
Don't use when:
- The diff is a one-line typo in a comment or doc
- The change is in a file with no AST coverage (e.g. binary, image, .lock)
- You only need to describe the diff, not judge it → use plain
git diff
Why this beats a generic LLM diff-read
| Concern | LLM-only diff read | tsa-pr-review |
|---|---|---|
| Classify signature vs body change | Heuristic | edit action=pr AST diff |
| Exact tests to run | Invented / hallucinated | edit action=impact verification_command |
| Caller blast radius | Grep guess | nav action=callers from persisted graph |
| Callee regression surface | Read N files | nav action=callees from persisted graph |
| Architecture rule violations | None | edit action=constraints vs YAML rules |
| Per-file pre-merge verdict | Subjective | edit action=safe SAFE/REVIEW/CAUTION/UNSAFE |
| Behaviour on Python 3.14 | n/a | Stable (grep-ast crashes, cgc silent-0) |
(Source: docs/internal/COMPETITOR_HEAD_TO_HEAD_2026-05-23.md — the 4
fault-tolerance advantages: oracle-consistent AST, indexed-not-silent,
Python 3.14-safe, no external DB dependency.)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 288 lines · 354 tokens per session scan A 110c6530a359
tsa-pr-review is a skill published in the GitHub repository aimasteracc/tree-sitter-analyzer (49 stars, last pushed today), licensed MIT. It adds 354 tokens to every session and 2,980 once invoked, about $0.0018 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
memstack-development-test-writer
Use this skill when the user says 'write tests', 'add tests', 'test coverage', 'unit tests', 'integration tests', 'component tests', 'mocking', 'edge cases', or needs to generate tests with proper mocking and edge case coverage. Do NOT use for refactoring plans or database migrations.
coverage
Compute code coverage for active track or module. Targets 95%+ coverage with report and justification for uncovered lines. Complements TDD workflow.
improving-tests
Improve test design, speed, and coverage with behavior-focused tests, useful seams, characterization tests, TDD, and test refactoring. Use when improving tests, optimizing slow suites, adding coverage, refactoring brittle tests, removing test waste, or working test-first. NOT for fixing production bugs (use…
write-unit-tests
Write comprehensive unit tests with proper setup, assertions, and coverage of happy paths, edge cases, and error scenarios.
write-tests
Write tests for existing production code. Processes ONE file at a time through a full pipeline: analyze, inventory (frozen BEFORE writing), write, executable coverage gate, verify, blind coverage audit, adversarial review, log. Uses CodeSift for discovery and analysis when available. Modes: [path] (specific target)…