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 commands/axiomantic/spellbook/dedupe-setupgit 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 | $0.00042 | $0.02442 |
| Opus 5 | $0.00021 | $0.01221 |
| Sonnet 5 | $0.00008 | $0.00488 |
| Haiku 4.5 | $0.00004 | $0.00244 |
Grade A, and why
dedupe-setup 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 — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MISSION
Phase 1 of the dedupe skill: turn an operator-supplied seed (file paths or
directories of markdown) into a deterministic, reviewable blocks manifest
that downstream phases consume.
This command is orchestrator-only. No subagents are dispatched here. All work is local filesystem inspection, segmentation, and manifest emission.
Part of the dedupe- command family.* Run before /dedupe-analyze.
Invariant Principles
- Refuse on bad input — every seed path must exist. A missing path aborts the run with an explicit error and emits no manifest.
- Resolve canonically — symlinks and relative paths are reduced to their canonical absolute form before deduplication.
- Respect
.gitignoreby default — gitignored files are excluded unless the operator passes--include-gitignored. - Segmentation is delegated — block boundaries and the bucket key /
finding_idderivation are defined inskills/dedupe/references/segmentation-protocol.md. This command applies that protocol; it does not restate it. - Mechanical safety floor is applied at setup time — the pattern
table in
skills/dedupe/references/safety-markers.mdis matched against every block body, and matching blocks are force-markedinline_mandatory_mechanical=truebefore any later phase runs. - No Python, no shell scripts — POSIX utilities only (
grep,awk,sed,shasum,printf,find,git ls-files), invoked via the harness Bash tool.
Safety-floor invariant: The mechanical safety floor runs as part of setup. Blocks flagged by the safety-marker predicate carry that flag forward through every later phase. The LLM safety screen in analyze may only ADD INLINE-MANDATORY status to a block, never remove mechanically-assigned status.
Invocation
/dedupe-setup <seed-path-or-comma-list> [--include-gitignored]
<seed-path-or-comma-list>— one or more paths separated by commas. Each entry may be a single markdown file or a directory. Directories are walked recursively for*.mdfiles.--include-gitignored— opt-in flag; when set,.gitignoredfiles under the seed are included in the manifest.
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 · 276 lines · 42 tokens per session scan A dc6ec6a48b72
dedupe-setup is a command published in the GitHub repository axiomantic/spellbook (10 stars, last pushed 7d ago), licensed MIT. It adds 42 tokens to every session and 2,442 once invoked, about $0.0002 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 commands, from other repositories
audit-agents-skills
Audit quality of agents, skills, and commands in a Claude Code project.
audit-whitepapers
Audit version freshness, FR/EN parity, and metadata quality of all whitepapers and recap cards.
audit-repo-docs
Audit repository documentation against 85+ best practices from claude-code-ultimate-guide.
boldguy-linkedin-answer
Draft a LinkedIn comment on a post, referencing the guide's relevant section, in Flow style with AI markers removed.
methodology-advisor
Analyzes your codebase and asks 3 targeted questions to recommend the right AI-assisted development methodology stack.
release
Release guide version (CHANGELOG + VERSION + README + landing sync + commit + push).