Menu
Output formats, -D/-S section discovery and selection, value projection, @ categories, and output limits shared across all commands.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Find evidence instead of guessing for .NET packages, platform libraries, local assemblies, APIs, dependencies, and version-to-version API changes.
Whole-assembly call-graph leverage ranking and performance triage for libraries (experimental).
Decide whether a change or upgrade is safe — API-surface diffs, behavioral diffs (allocations, exceptions), feature switches, and version resolution.
Judge whether code is sound and safe to call — its exception surface (throws, catches, exception types) and the unsafe operations in a method body.
Reconstruct a method or type locally as C# and IL — decompiled source, annotated source with hidden facts, raw IL, fidelity levels, and IL-offset lookup. Always local, no network.
Use for opt-in Deep Inspect run triage: test lane failures, census artifacts, pinned-subset drift, and routing findings into issues or tracker updates.
| name | dotnet-inspect-query |
| version | 0.1.0 |
| description | Output formats, -D/-S section discovery and selection, value projection, @ categories, and output limits shared across all commands. |
The query system is like Go templates, without a DSL: every command emits the
same structured sections, and you discover, select, and project them with the
same cross-command flags — on find, type, member, package, library,
diff, and the relationship commands. Discover the shape first, then select and
project.
dnx dotnet-inspect -y -- <command>
Default output is Markdown. Pick a machine or compact shape when you need one:
--table — compact aligned rows.--tsv — stable snake_case headers, no embedded tabs/newlines.--jsonl — one JSON object per row.--json-array — one JSON array for projected rows (--urls, --paths, --value, --print-all).--json — structured documents.--bare — one undecorated payload or URL list.--count — a bare row count.--value / --urls / --paths — project one selected section to scalar, URL, or path payloads.--print — print one document behind a selected printable row; use --row N when multiple printable rows exist.--print-all — print every printable row from one selected section, with text separators, --jsonl, or --json-array.--mermaid — graph-shaped output.Use -D to discover the sections and columns a command can emit, -S Section to
select sections by name or wildcard, and --columns/--fields to project
values. Discover first instead of guessing names.
dnx dotnet-inspect -y -- member JsonSerializer --platform System.Text.Json -D --tsv
dnx dotnet-inspect -y -- member JsonSerializer --platform System.Text.Json -m Serialize -D "Member Index" --tsv
dnx dotnet-inspect -y -- member JsonSerializer --platform System.Text.Json -m Serialize -S "Member Index" --columns "Selector;Stable;Canonical Signature" --tsv
@ names a category of sections: -S @All, -S @Source, -S @Audit,
-S @Integrations, -S @Switches. Row formats (--tsv/--jsonl/--table)
work best with one concrete section, not a category.
Prefer built-in limits to shell pipes:
-n N and numeric shorthand like -6 cap output lines, like head.--tail N shows the end, like tail.--rows makes -n cap Markdown table data rows instead of output lines.--print, --value, --urls, or --paths, --row N chooses the projected row; -n N still limits output lines.--count counts rows in one selected table.Command-specific caps: -t N for type/find rows, -m N for members, and
--versions N for package versions.