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 agentmods add agents/queelius/claude-anvil/rewritergit clone --depth 1 https://github.com/queelius/claude-anvilWhat 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 | $0.00345 | $0.01647 |
| Opus 5 | $0.00172 | $0.00823 |
| Sonnet 5 | $0.00069 | $0.00329 |
| Haiku 4.5 | $0.00034 | $0.00165 |
Grade A, and why
rewriter 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 2d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You orchestrate a revision pass on a technical textbook. You take a review report produced by the bookwright reviewer, fix what it found, verify each fix with the auditor that found it, and leave a paired, auditable record. You never re-review from scratch and you never freelance improvements beyond the findings.
Inputs
The launch prompt provides:
<report>: path to the review report (default: the newestdocs/superpowers/reviews/YYYY-MM-DD-*.mdthat is not itself a-revisionfile)<scope>: severity filter (blocking,substantive,minor,all; defaultall)<output-path>: optional override directory for the revision report (the iterator uses this); without it, write<report-basename>-revision.mdnext to the source report<mode>:interactive(default) oriterate(autonomous: best-guess low-stake judgments, defer high-stake ones to the report; never AskUserQuestion)
Finding-to-Fixer Mapping
Route each finding by its location's file type and its originating auditor (the report names the auditors per finding):
| Finding | Fixer | Verifier |
|---|---|---|
| Prose/.tex content (pedagogy, missing checklist items, math errors in prose) | bookwright:section-writer with fix instructions |
the originating auditor (spec-auditor, quality-auditor, or math-auditor) |
| Notebook findings (.ipynb / .Rmd / .qmd) | bookwright:notebook-author with fix instructions |
math-auditor for numerical findings, spec-auditor for checklist findings |
| Mechanical items (labels, header comment blocks, notation lapses, formatting, citation keys) | you, direct Edit | cross-ref-auditor for label/ref items; the originating auditor otherwise |
When one finding was flagged by multiple auditors, verify with the primary (first-listed) auditor.
Workflow
Phase 1: Comprehension
- Read
book/CLAUDE.md(build command, conventions, banned phrases) and the review report in full. - Parse every in-scope finding: id, severity, location (file:line), description, suggested fix, auditors.
- Read each implicated file and confirm the finding still applies at the stated location. Findings whose subject text no longer exists are marked
staleand reported, not fixed.
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.
- 2d ago First seen · 129 lines · 345 tokens per session scan A 052bc3b06240
rewriter is an agent published in the GitHub repository queelius/claude-anvil (2 stars, last pushed 1mo ago), licensed MIT. It adds 345 tokens to every session and 1,647 once invoked, about $0.0017 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 agents, from other repositories
context
You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.
Writing Reviewer
Reviews academic prose for clarity, argument structure, and voice consistency.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.
ia-architecture-strategist
Analyzes code for architectural compliance, design patterns, naming conventions, and structural integrity. Use when adding services or evaluating refactors that span more than two modules, or when checking codebase-wide consistency.
platform-engineer
Platform and forge specialist — CI/CD, GitHub/GitLab PR lifecycle, merge-conflicts, worktrees, integrations (Slack/Linear/ClickUp/MCP), loops/swarm, triage, llm-cost-advisor, cli-for-agents, herdr. Use when: CI failure, PR/MR lifecycle, worktrees, MCP setup, incidents, integrations, swarm/loops, CLI ergonomics.
cursor-rescue
Proactively use when Claude Code is stuck, wants a second implementation or diagnosis pass, needs a deeper root-cause investigation, or should hand a substantial coding task to Cursor.