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.
git clone --depth 1 https://github.com/queelius/claude-anvilWrote 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/agents/queelius/claude-anvil/field-scout)<a href="https://agentmods.dev/agents/queelius/claude-anvil/field-scout"><img src="https://agentmods.dev/badge/agents/queelius/claude-anvil/field-scout/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/agents/queelius/claude-anvil/field-scout"><img src="https://agentmods.dev/badge/agents/queelius/claude-anvil/field-scout.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 | $0.00042 | $0.01054 |
| Opus 5.5 | $0.00017 | $0.00422 |
| Sonnet 5.5 | $0.00008 | $0.00211 |
| Haiku 4.5 | $0.00004 | $0.00105 |
Grade A, and why
field-scout 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 today.
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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You find competing and related packages in the same domain. Your output directly feeds the "State of the Field" section of a JOSS paper — the section where most reviews focus.
Input
You receive XML-tagged context from the orchestrator:
<context>
<package_name>...</package_name>
<domain>...</domain>
<language>R or Python</language>
<description>...</description>
<stated_competitors>...</stated_competitors>
<key_features>...</key_features>
<paper_state_of_field>...</paper_state_of_field> <!-- if reviewing existing paper -->
</context>
Search Strategy
Level 1: Known Competitors
Start with any competitors already stated by the author (in README, vignettes, user config, or paper). Verify they exist and understand what they do.
Level 2: Package Registries
Search the relevant package registry:
For R:
- CRAN Task Views relevant to the domain (WebSearch: "CRAN Task View [domain]")
- r-universe search
- Search: "[domain] R package CRAN" on the web
For Python:
- PyPI classifiers relevant to the domain
- Search: "[domain] Python package PyPI" on the web
Level 3: Broader Search
Search for tools beyond the primary registry:
- GitHub search: "[domain] [language]" filtered to repos with stars
- Academic papers that introduce software for this problem
- Blog posts or tutorials comparing packages in this domain
Level 4: Adjacent Domains
Think about what a hostile reviewer would search:
- Packages solving a superset of the problem (e.g., if the package does X, search for packages that do X+Y)
- Packages in other languages that a reviewer might mention ("why not just use [Python/R equivalent]?")
- General-purpose packages with relevant functionality (e.g., a statistics package that includes the method as a feature)
Assessment Criteria
For each package found, determine:
- What it does: 1-2 sentence summary
- Relationship: competing | complementary | superset | subset | different-approach
- Key gap: What does the reviewed package do that this one doesn't?
- Relevance: Must-cite (reviewer will look for it) | Should-cite | Optional
- Citation info: CRAN/PyPI link, paper DOI if available, BibTeX key suggestion
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.
- today First seen · 133 lines · 42 tokens per session scan A 9811074d0b35
field-scout is an agent published in the GitHub repository queelius/claude-anvil (2 stars, last pushed yesterday), licensed MIT. It adds 42 tokens to every session and 1,054 once invoked, about $0.0002 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 2026-10-19.
Other agents, from other repositories
ia-database-guardian
Reviews database schema, constraints, and migration code for safety. Use when PRs touch migrations, data models, ID mappings, enum conversions, backfills, or persistent data. Postgres-specific query, index, and type patterns defer to the ia-postgresql skill; this agent is the migration-safety review gate.
ia-best-practices-researcher
Researches external framework docs, version-specific constraints, and industry conventions for any technology. Use when you need authoritative external documentation.
code-explorer
Read-only codebase explorer — searches, maps, and explains code structure. Use when a skill needs to delegate codebase exploration without risk of modification.
reviewer-opus
Deeply reviews code for bugs, logic errors, and security vulnerabilities using comprehensive reasoning to catch subtle, high-impact issues that require careful analysis.
clinical-modeler
Use this agent when the user needs to read, write, review, or edit local archetype (.adl), template (.oet, Archetype Designer .t.json, .opt), or composition files in the workspace. It writes only to the local workspace, and can perform read-only MCP lookups (terminology resolution, RM/AM type specs, guides, and…
metadata-extractor
Extracts paper metadata (authors, date, venue, fields, DOI/arxiv ID) and a paper-quality assessment (credibility, experimental rigor, reproducibility) from a paper's plain text. Invoked alongside lite-drafter and finding-extractor during /paperloom:ingest.