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 root-causegit 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/root-cause)<a href="https://agentmods.dev/skills/larsboes/axon/root-cause"><img src="https://agentmods.dev/badge/skills/larsboes/axon/root-cause/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/root-cause"><img src="https://agentmods.dev/badge/skills/larsboes/axon/root-cause.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 | $0.00166 | $0.01417 |
| Opus 5 | $0.00083 | $0.00709 |
| Sonnet 5 | $0.00033 | $0.00283 |
| Haiku 4.5 | $0.00017 | $0.00142 |
Grade A, and why
root-cause 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 9d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
root-cause
One failure goes in. A named cause, the probe that would falsify it, and a fix ranked by whether it removes the class come out.
Root-cause explains a failure that already happened. To attack a proposal before it ships, use
red-team. To choose between two repairs, use council.
Adapted from the RootCauseAnalysis skill in LifeOS by Daniel Miessler (https://github.com/danielmiessler/LifeOS), MIT.
Procedure
Copy this checklist and track progress:
- [ ] 1. State the failure as an observation, not a diagnosis
- [ ] 2. Build the timeline
- [ ] 3. List what changed
- [ ] 4. Pick the method
- [ ] 5. Generate candidate causes, then try to refute each one
- [ ] 6. Name the cause and its falsifying probe
- [ ] 7. Rank the fixes by what class each one removes
1 — State the failure. Write what was observed, when it started, and how it was noticed. "The automations stopped running" is an observation. "The restart wiped them" is a diagnosis, and writing one here decides the investigation before it starts. Say which of the two you were handed.
2 — Build the timeline. Read the logs, the history, the git log, the state file — whatever records time. Put every dated fact in order, including the boring ones. Mark the last moment the system is known to have worked and the first moment it is known to have failed. The cause is between those two marks, and most investigations that fail never drew them.
3 — List what changed. In that window: deploys, config edits, upgrades, credentials, network, hardware, and anything a person or an agent ran by hand. A failure with no change behind it is possible but rare; when the list is empty, say so explicitly, because it moves the answer toward a latent condition that was always there and toward what made it visible today.
4 — Pick the method. Read references/methods.md and pick the one the evidence supports:
| The evidence you have | Method |
|---|---|
| A clean timeline and a suspect change | Change analysis |
| One failure, a chain of mechanism to walk back | Five whys |
| It fails here and not there, now and not then | Is / is-not |
| Several possible paths to the same symptom | Fault tree |
| It already failed more than once | Recurrence pattern |
| An incident that is over and needs a record | Blameless postmortem |
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 106 lines · 166 tokens per session scan A 9c8b9808a9ee
root-cause is a skill published in the GitHub repository larsboes/Axon (1 stars, last pushed yesterday), licensed MIT. It adds 166 tokens to every session and 1,417 once invoked, about $0.0008 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 2026-09-07.
Other skills, from other repositories
build-test
Run the project's build / typecheck / lint / test commands and emit the build.passing + tests.passing signals devloop convergence reads.
patch-edit
Apply one rewrite-plan step at a time as small reviewable file edits, never rewriting whole files when a localised change suffices.
cvm-ai-doctor
A health-diagnosis workflow for servers, computers, virtual machines, and containers on Linux, macOS, or Windows.
gini-bug-report
File a locally-captured, already-redacted Gini crash report as a GitHub issue, with the user's consent. Reads the pending crash queue and delegates the actual filing to the github-issues skill.
fix-finding
Use when the user explicitly asks to fix and verify a validated or plausible security finding. Do not use as the primary trigger for full PR, commit, branch, patch, or repository scans.
compare-files
Compare two files and show differences.