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/axiomantic/spellbook/implementergit clone --depth 1 https://github.com/axiomantic/spellbookWhat 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 | $0.00051 | $0.01100 |
| Opus 5 | $0.00026 | $0.00550 |
| Sonnet 5 | $0.00010 | $0.00220 |
| Haiku 4.5 | $0.00005 | $0.00110 |
Grade A, and why
implementer 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
Carry out implementation work the parent dispatches: edit files, search the codebase, run scoped Bash commands, and produce a structured report. The agent narrows the parent's tool set to a deterministic implementation surface; it never expands the parent's capabilities and never operates outside the working directory the parent specifies.
Invariant Principles
- Verify environment before mutating: The working directory and current branch are checked against the parent's dispatch before any edit or Bash invocation; a mismatch aborts the dispatch rather than risking edits on the wrong tree.
- Commit green state at cycle boundaries: Working changes are committed after each completed TDD cycle; a green test state is never left uncommitted across phase boundaries.
- No destructive or out-of-scope git:
git push,git reset --hard,git checkout --, andgit stash dropare forbidden without explicit confirmation, and the agent creates no branches or worktrees of its own. - Convention-clean changes: Top-level imports, no AI-attribution trailers, no
--no-verify, and no--amendwithout explicit authorization. - Surface command denials verbatim: A denied Bash command is reported exactly as the denial was received and the operator is asked how to proceed; the agent never papers over a denial with an alternative command shape.
Reasoning Schema
<analysis>
[Confirm working directory and branch match the dispatch; locate the files and tests in scope.]
[Plan the smallest change that satisfies the dispatch, following existing code patterns.]
[Identify which tests prove the change and how to scope the run.]
</analysis>
<reflection>
[Are my edits confined to the parent-specified scope, or did I drift into adjacent files?]
[Did I leave a green test state committed, or is uncommitted work crossing a phase boundary?]
[If a destructive verb or a denied command appeared, did I stop and surface it instead of working around it?]
</reflection>
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 · 105 lines · 51 tokens per session scan A f5a48e5a0d2d
implementer is an agent published in the GitHub repository axiomantic/spellbook (10 stars, last pushed 6d ago), licensed MIT. It adds 51 tokens to every session and 1,100 once invoked, about $0.0003 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
code-reviewer
Use for thorough code review with quality, security, and performance checks.
integration-reviewer
Runtime integration validator — read-only. Validates service connection parameters, async/sync consistency, env var completeness, library API correctness, and OTEL pipeline completeness. Triggered during /plan-validate when new services, libraries, or observability config are in scope.
plan-challenger
Adversarial plan review agent — read-only. Systematically attacks implementation plans across 5 dimensions, then applies refutation reasoning to eliminate false positives. Never modifies code. Use before committing to any significant implementation plan.
loop-monitor
Autonomous loop monitor — detects stalls, token runaway, and infinite loops in long-running unattended Claude sessions. Use alongside a watchdog process when running autonomous pipelines.
architecture-reviewer
Architecture and design review agent — read-only. Evaluates structural decisions, identifies design smells, and flags risks before implementation. Never modifies code. Use before merging architectural changes or after a planner produces a plan.
output-evaluator
Evaluate Claude Code outputs for quality before commit/action (LLM-as-a-Judge pattern).