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 skills add larsboes/Axon --skill divergegit clone --depth 1 https://github.com/larsboes/AxonWrote 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/skills/larsboes/axon/diverge)<a href="https://agentmods.dev/skills/larsboes/axon/diverge"><img src="https://agentmods.dev/badge/skills/larsboes/axon/diverge/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/skills/larsboes/axon/diverge"><img src="https://agentmods.dev/badge/skills/larsboes/axon/diverge.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 | $0.00156 | $0.01270 |
| Opus 5.5 | $0.00062 | $0.00508 |
| Sonnet 5.5 | $0.00031 | $0.00254 |
| Haiku 4.5 | $0.00016 | $0.00127 |
Grade A, and why
diverge 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 1mo 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
diverge
Every other capability in this pack needs options to already exist. council needs two or more,
red-team needs one proposal, crystallize needs notes, root-cause needs a failure. Nothing
produces candidates. This is that missing half, and it is deliberately the only skill here that
generates rather than judges.
Generating is the easy half and the model does it freely. The work is breadth that is real and then winnowing, which is why this skill spends most of its words on the second.
Procedure
- [ ] 1. State what the candidates are candidates FOR
- [ ] 2. Name the generators, one per candidate
- [ ] 3. Produce the pile, one candidate per generator
- [ ] 4. Winnow with `narrow`
- [ ] 5. Hand the survivors on
1 — State the target. A pile without a question is a list. Write the decision or the goal the
candidates are meant to serve, and if the caller gave a topic rather than a question, confirm the
restatement with ask first — council's references/compose.md says how, and the same reasoning
applies: a pile generated against the wrong target is entirely wasted work.
2 — Name the generators before producing anything. Pick five to eight and write them down. The list is what makes the breadth checkable afterwards:
| Generator | The question it asks |
|---|---|
| Analogy | What solved a structurally similar problem somewhere else? |
| Inversion | What if we did the opposite, or optimised for the thing we are avoiding? |
| Constraint removal | Which constraint, if it vanished, would make this easy? |
| Extreme scale | What would this look like at 100x, or at 1/100th? |
| Recombination | Which two things we already have, combined, answer this? |
| Elimination | What if we removed the part everyone assumes is required? |
| Precedent | What did this repository already try and abandon, and why? |
3 — Produce one candidate per generator. One. Not five from analogy and one from everything else, which is what happens when the first idea is good. If a generator yields nothing, say so and name it — an empty generator is a finding, and it is often the most interesting line in the pile.
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.
- 1mo ago First seen · 86 lines · 156 tokens per session scan A 4774f1310bf2
diverge is a skill published in the GitHub repository larsboes/Axon (1 stars, last pushed 1mo ago), licensed MIT. It adds 156 tokens to every session and 1,270 once invoked, about $0.0006 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 2026-09-23.
diverge compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| divergelarsboes/Axon | Skill | 1 | 1mo ago | MIT | A | 156 | Claude Code |
| ligandmpnnaipoch/open-science | Skill | 5.5k | today | Apache-2.0 | A | 100 | Claude Code, Codex |
| evo2aipoch/open-science | Skill | 5.5k | today | Apache-2.0 | A | 83 | Claude Code, Codex |
| create-agentPM-Shawn/Abu-Cowork | Skill | 419 | 3d ago | Apache-2.0 | A | 13 | Claude Code |
| content-monitorCoWork-OS/CoWork-OS | Skill | 537 | today | MIT | A | 16 | Claude Code, Codex |
| llm-wikiCoWork-OS/CoWork-OS | Skill | 537 | today | MIT | A | 27 | Claude Code, Codex |
| simplifyCoWork-OS/CoWork-OS | Skill | 537 | today | MIT | A | 24 | Claude Code, Codex |