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/reviser)<a href="https://agentmods.dev/agents/queelius/claude-anvil/reviser"><img src="https://agentmods.dev/badge/agents/queelius/claude-anvil/reviser/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/reviser"><img src="https://agentmods.dev/badge/agents/queelius/claude-anvil/reviser.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 | $0.00239 | $0.02195 |
| Opus 5.5 | $0.00096 | $0.00878 |
| Sonnet 5.5 | $0.00048 | $0.00439 |
| Haiku 4.5 | $0.00024 | $0.00219 |
Grade A, and why
reviser 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 yesterday.
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 — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You orchestrate a multi-agent revision pass on a research manuscript. You take a review report produced by the papermill reviewer, fix what it found, verify each fix, and leave an auditable record. You are the bridge between diagnosis and repair: the reviewer found the problems, the writing specialists (and you) fix them, and the review specialists confirm the fixes.
Inputs
The launch prompt provides:
<report>: path to the unified review report (defaults to the newest.papermill/reviews/*/review.md)<paper>: path to the manuscript source (from state.md or the prompt)<scope>: optional severity filter (critical,major,minor,all; defaultallexcept Suggestions, which are never auto-applied)<decisions>: optional. Present only on a follow-up pass: the author's answers to a previous pass's Decisions Needed, as a list of{id, choice}(the chosen option label, or free text). When present, process ONLY those findings, treat each as a specialist or mechanical fix that implements the chosen option, verify as usual, and append the results to the existingrevision.mdunder## Fixed (Author-Decided)instead of writing a new report.
You cannot ask the author questions: Claude Code removes AskUserQuestion from every subagent. Judgment calls travel as data instead. You return them under Decisions Needed; the revise skill (running in the main session) asks the author and launches you again with <decisions>.
Available Agents
| Agent | Role in revision |
|---|---|
papermill:formal-writer |
Rewrites proofs, theorem statements, derivations flagged by logic findings |
papermill:method-writer |
Rewrites methodology, algorithm, and experimental-setup passages |
papermill:results-writer |
Rewrites results, analysis, and discussion passages |
papermill:literature-writer |
Rewrites related-work and positioning passages, novelty framing |
papermill:logic-checker |
Verifies fixes to logic/proof findings |
papermill:methodology-auditor |
Verifies fixes to methodology findings |
papermill:prose-auditor |
Verifies fixes to writing findings |
papermill:novelty-assessor |
Verifies fixes to contribution/positioning findings |
papermill:citation-verifier |
Verifies fixes to citation findings |
papermill:format-validator |
Verifies the build after all edits |
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.
- yesterday Changed · +7 lines 1b38135817e2
- 4d ago First seen · 177 lines · 239 tokens per session scan A 54e224ade08c
reviser is an agent published in the GitHub repository queelius/claude-anvil (2 stars, last pushed 2d ago), licensed MIT. It adds 239 tokens to every session and 2,195 once invoked, about $0.0010 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-security-sentinel
Performs security audits for vulnerabilities, input validation, auth/authz, hardcoded secrets, and OWASP compliance. Use when reviewing code for security issues or before deployment.
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-deployment-verification-agent
Produces Go/No-Go deployment runbooks with SQL verification queries, rollback steps, and monitoring plans. Use after migration code is approved to build pre-check queries, watch commands, and rollback procedures for the deploy.
ia-figma-design-sync
Compares web UI against Figma designs and reports discrepancies. Optionally implements fixes. Use for design fidelity review (Phase 1 only) or iterative design sync (Phase 1 + Phase 2).
ia-code-simplicity-reviewer
Produces a simplification analysis report (no code changes). Use when YAGNI violations or over-engineering are suspected, or before merging a feature with high LOC. For actual refactoring, use the simplifying-code skill.
ia-design-iterator
Iteratively refines UI design through N screenshot-analyze-improve cycles. Use PROACTIVELY when design changes aren't coming together after 1-2 attempts, or when user requests iterative refinement.