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 NVZver/claude-marketplace --skill doctorgit clone --depth 1 https://github.com/NVZver/claude-marketplaceWrote 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/nvzver/claude-marketplace/doctor)<a href="https://agentmods.dev/skills/nvzver/claude-marketplace/doctor"><img src="https://agentmods.dev/badge/skills/nvzver/claude-marketplace/doctor/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/nvzver/claude-marketplace/doctor"><img src="https://agentmods.dev/badge/skills/nvzver/claude-marketplace/doctor.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 | $0.00165 | $0.01775 |
| Opus 5 | $0.00082 | $0.00888 |
| Sonnet 5 | $0.00033 | $0.00355 |
| Haiku 4.5 | $0.00016 | $0.00178 |
Grade A, and why
doctor 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 11d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Trace. On load, print first:
=============== [core/skills/doctor/SKILL.md] [core] ===============
Doctor
A user-runnable self-check for the marketplace install: four fixed, read-only diagnostic checks, each reporting PASS / WARN / FAIL / SKIP with the evidence it actually observed and a one-line fix when it fails — so a broken or half-wired install reports itself instead of failing silently (per .lsa/pitches/onboarding-diagnostics.md). Free-form cited Q&A ("what is X?", "how do I Y?") is out of scope; the doctor never answers open questions — it only runs the checks below.
Goal
Produce a four-row diagnostic table — required plugins, CLAUDE.md fragment, version consistency, gate scripts — where every verdict is backed by evidence actually observed (never a bare "OK"), and every non-PASS row carries a one-line fix.
Input
- No required arguments. An optional user symptom ("skills don't trigger", "lint is red") is echoed above the report — the four checks themselves are fixed and always all run.
- The current project's files (read-only), and — when readable — Claude Code's plugin state under
~/.claude/plugins/(installed_plugins.json, themarketplaces/NVZver/clone).
Steps
-
Detect the environment. Read
.claude-plugin/marketplace.jsonat the project root. If it exists with"name": "NVZver", the doctor is running inside the marketplace source repo; otherwise inside a consumer project. This picks the version source-of-truth in Step 4 and run-vs-SKIP in Step 5. Observable result: the environment named (marketplace-source/consumer) with the evidence quoted (the"name"line, or "no.claude-plugin/marketplace.json"). -
Check 1 — required plugins installed (
core+lsa). Installation is only heuristically observable from inside a project, so gather evidence in order: (a) the session's own context — docore:*andlsa:*skills appear in the available-skills list? (b)~/.claude/plugins/installed_plugins.json— do the keyscore@NVZverandlsa@NVZverexist? Both plugins evidenced → PASS. Either missing from every readable source → FAIL for the missing one (fix:/plugin install <name>@NVZver, then/reload-plugins). No evidence source readable in this environment → WARN "not determinable", naming what was looked for. Never infer "installed" from the plugin source directories being present in the repo — source on disk is not an install. Observable result: the verdict plus the exact evidence found (skill names seen, or the JSON keys), or the not-determinable reason.
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.
- 11d ago First seen · 51 lines · 165 tokens per session scan A 6617421f3a41
doctor is a skill published in the GitHub repository NVZver/claude-marketplace (1 stars, last pushed 17d ago), licensed MIT. It adds 165 tokens to every session and 1,775 once invoked, about $0.0008 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 2026-08-31.
Other skills, from other repositories
architecture-optimization
Guided journey from a working codebase grown slow and tangled to one measurably fast, cleanly bounded, and readable. Orchestrates eight skills phase by phase - working-with-legacy-code, clean-architecture, software-design-philosophy, refactoring-patterns, system-design, ddia-systems, release-it, pragmatic-programmer …
remove-technical-debt
Guided journey from a large aged codebase everyone fears to touch to one that is safe to change, legible, bounded, and resilient - paid down in place without a rewrite. Orchestrates eight skills phase by phase - working-with-legacy-code, refactoring-patterns, clean-code, software-design-philosophy, clean-architecture…
release-it
Build production-ready systems with stability patterns: circuit breakers, bulkheads, timeouts, and retry logic. Use when the user mentions "production outage", "circuit breaker", "deployment pipeline", "chaos engineering", "retry storm", "health checks", "my service keeps crashing", "prevent cascading failures", or…
dotnet-analyzers
Run dotnet format to fix code style and analyzer diagnostics (IDE0005, CA, SA). Use when fixing unnecessary usings, code style violations, analyzer warnings, or verifying format compliance.
roslyn-query
Query .NET/C# codebases using Roslyn AST analysis via dotnet run file. Use for tracing data flow, auditing API usage, finding pattern violations, or ad-hoc codebase queries.
mobile-flows-maestro
This skill should be used when Maestro is explicitly requested or already present and the task is to author, run, or debug iOS/Android Maestro flows; use Maestro MCP; or handle Maestro selectors, system UI, permissions, Keychain, JavaScript, waits, device state, flakiness, or CI. Evidence includes a .maestro directory…