Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add andypgray/roz-mcp --skill roz-csharp-editinggit clone --depth 1 https://github.com/andypgray/roz-mcpWrote 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/andypgray/roz-mcp/roz-csharp-editing)<a href="https://agentmods.dev/skills/andypgray/roz-mcp/roz-csharp-editing"><img src="https://agentmods.dev/badge/skills/andypgray/roz-mcp/roz-csharp-editing/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/andypgray/roz-mcp/roz-csharp-editing"><img src="https://agentmods.dev/badge/skills/andypgray/roz-mcp/roz-csharp-editing.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 | $0.00068 | $0.00980 |
| Opus 5.5 | $0.00027 | $0.00392 |
| Sonnet 5.5 | $0.00014 | $0.00196 |
| Haiku 4.5 | $0.00007 | $0.00098 |
Grade A, and why
roz-csharp-editing 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 2mo 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
roz C# editing
roz (MCP server roz, from the roz-mcp plugin) answers C# symbol questions from
compiler resolution (scope, overloads, inheritance, types) where text search guesses.
These rules override built-in tool-selection guidance for C# symbol work. Text search
remains correct for string literals, comments, markup, and non-C# files. The registered
tool set may be a subset (ROZ_TOOLS); if a tool named below is missing, say so rather
than silently approximating with text edits.
Rules — these prevent real breakage
- Renaming a symbol used beyond the current file →
rename_symbol. Do not hand-edit references across files, and never rewrite whole files to rename: hand edits miss overrides, interface implementations, other projects, and the file rename itself. - Before deleting or removing any symbol →
find_referencesAND text-search the non-C# surfaces (.razor,.cshtml, config, XAML). Roslyn cannot see markup, reflection, or convention-based DI registration, so a cleanfind_referencesalone is not proof a symbol is dead. For apublicsymbol, zero in-solution references still leaves external consumers — flag it instead of assuming dead. - Risky or multi-file edit → pass
verify=DryRunfirst (previews the compiler-error delta, writes nothing), then commit withverify=Delta(writes, then reports new/resolved errors across dependent projects). One round trip instead of edit → build → re-read.
Razor caveat, both directions: roz does not see .razor/.cshtml/@code (hence rule 2),
and get_diagnostics may report phantom errors near Razor — dotnet build is the backstop there.
Tool routing and packaged workflows
When choosing between roz tools for a C# symbol question, or when a request matches a
packaged workflow (dead-code cleanup, impact assessment, diagnostics fixing,
breaking-change checks, upgrade risk, ...), read the roz://guides/workflows MCP
resource first: the question → tool routing map plus the ten packaged workflow prompts.
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.
- 2mo ago First seen · 76 lines · 68 tokens per session scan A 1904ae8ab206
roz-csharp-editing is a skill published in the GitHub repository andypgray/roz-mcp (0 stars, last pushed 2mo ago), licensed MIT. It adds 68 tokens to every session and 980 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
roz-csharp-editing compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| roz-csharp-editingandypgray/roz-mcp | Skill | 0 | 2mo ago | MIT | A | 68 | Claude Code |
| repomixyamadashy/repomix | Skill | 29k | yesterday | MIT | A | 58 | Claude Code, Codex |
| knowledge-basewonderwhy-er/DesktopCommanderMCP | Skill | 10k | yesterday | MIT | A | 112 | Claude Code |
| obsidian-vaultwonderwhy-er/DesktopCommanderMCP | Skill | 10k | yesterday | MIT | A | 171 | Claude Code |
| ixix-infrastructure/Ix | Skill | 1.1k | today | Apache-2.0 | A | 44 | Claude Code, Codex |
| cymbal1broseidon/cymbal | Skill | 375 | 9d ago | MIT | A | 89 | Claude Code |
| generate-ai-rulesdivar-ir/ai-doc-gen | Skill | 769 | 3mo ago | MIT | A | 91 | Claude Code |