snippet-eval — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited snippet-eval (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You use `analyze_snippet` and `evaluate_csharp` for fast feedback when loading a full `.sln` is unnecessary or impossible.
$ARGUMENTS should indicate either:
Full solution workflows live in `roslyn://server/catalog`. This skill is for ephemeral analysis only.
| Goal | Tool |
|---|---|
| Parse, bind, diagnostics on a fragment | `analyze_snippet` |
| Run a script / evaluate an expression (side effects, output) | `evaluate_csharp` |
| Production refactor / cross-file impact | Load workspace with `workspace_load` and use symbol tools |
Invoke with --explain or ask "explain this snippet" / "what does this do?". The skill produces a step-by-step semantic walkthrough instead of a pass/fail result:
LocalReference, InvocationOperation, ConditionalAccessOperation)await, show the async flow: where the state machine suspends, what continuation happens on completion vs exception.Explain mode is read-only — it never calls evaluate_csharp. It's useful for teaching newcomers, debugging why a snippet doesn't compile, or understanding inherited code.
Treat `evaluate_csharp` as arbitrary code execution in the MCP host environment. Only run code the user explicitly supplied and avoid secrets in snippets. Explain mode is safe — it never executes.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.