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 ljucask/pureinn-product-development --skill pm-root-causegit clone --depth 1 https://github.com/ljucask/pureinn-product-developmentWrote 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/ljucask/pureinn-product-development/pm-root-cause)<a href="https://agentmods.dev/skills/ljucask/pureinn-product-development/pm-root-cause"><img src="https://agentmods.dev/badge/skills/ljucask/pureinn-product-development/pm-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/ljucask/pureinn-product-development/pm-root-cause"><img src="https://agentmods.dev/badge/skills/ljucask/pureinn-product-development/pm-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 | $0.00000 | $0.05217 |
| Opus 5 | $0.00000 | $0.02609 |
| Sonnet 5 | $0.00000 | $0.01043 |
| Haiku 4.5 | $0.00000 | $0.00522 |
Grade A, and why
pm-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 10d 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 — 357 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PM - Root Cause (diagnose an in-flight anomaly)
Agent mode (--agent)
This skill's value is the live dialogue - --agent is not supported. If invoked with --agent, warn once ("this skill needs interactive back-and-forth; agent mode would hollow it out") and proceed interactively.
What this skill does
Something live is behaving unexpectedly - a metric moved, users act strangely, churn spiked, a feature flopped. You describe the symptom; the skill runs a structured investigation and drills to the real root cause, then hands you testable hypotheses and the cheapest way to confirm.
This is in-flight diagnosis, not foundational validation. It differs from the neighbours:
pm-problem-validation- is the core problem real (at project start)?pm-stress-test- poke holes in my plan (adversarial, before a room)?pm-root-cause- why is THIS happening now (diagnostic, on a live thing)?
Two modes:
| Mode | When | What it does |
|---|---|---|
| Guided investigation (default) | You want to find the cause | Interactive Socratic drill through the steps: real vs artifact → localize → what changed → candidate causes → 5 Whys → evidence-vs-guess → confirm test |
| Quick differential | You just want the candidate-cause map | One pass: for the stated anomaly, the likely causes, their data signature, and exactly what data to pull to confirm/rule out - no long back-and-forth |
Data-optional (graceful). If you have analytics, the skill uses it. If not, it still structures the investigation and tells you exactly which data to pull and how to read it.
Hard rule: don't stop at the first plausible cause
The single most common failure of root-cause work is premature closure - grabbing the first believable answer and stopping. This skill actively resists it:
- Correlation is not causation. Two things moving together is a lead, not a cause. Demand the mechanism ("how exactly did A produce B?").
- Symptom is not root. "Engineer ran the wrong command" / "a bug" is a symptom - ask what allowed it.
- Rarely one cause. Complex systems usually need several contributing factors. Ask: if we removed this one factor, would the anomaly still happen?
- Separate what you KNOW (evidence) from what you GUESS (hypothesis). Never present a guess as a finding. Mark every candidate
[EVIDENCE]or[HYPOTHESIS - untested]. - Before accepting any root cause, run the validation checklist (end of skill).
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.
- 10d ago First seen · 357 lines · 0 tokens per session scan A c4e966f252bf
pm-root-cause is a skill published in the GitHub repository ljucask/pureinn-product-development (2 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 5,217 tokens. 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 skills, from other repositories
pr-babysitter
Use when monitoring GitHub pull request CI, diagnosing failed checks, rebasing branches onto github/main, applying narrowly scoped fixes, and updating PRs until their latest checks are green or a human blocker is identified.
submit-github-bug-issue
Use when converting QA findings, black-box failures, red-team reports, regression evidence, or local bug notes into GitHub Issues for NVIDIA/TensorRT-Model-Connect. Standardizes checking issue templates, checking labels, de-duplicating existing issues, drafting a bug report, creating the issue on GitHub, applying the…
operating-cadence
Designs the rhythm an organization runs on — which reviews happen weekly, monthly and quarterly, what each one decides, who owns the numbers presented, and how a signal at the front line reaches the people who can act on it. Use this to set up a management operating system, fix a meeting calendar that produces no…
debug-trt-mismatch
Diagnose a TensorRT-Model-Connect family whose native output disagrees with its declared reference or family-owned correctness contract.
profile-model
Measure one TensorRT-Model-Connect model through its public Task API or run a checked-in performance-matrix entry with comparable evidence.
debug-render
Debug a WRONG or imperfect render (not a hard error) by inspecting inputs and intermediate steps with run-to-node. Render one branch up to an output, preview-tap latents/masks/preprocessor maps, localize the first bad stage, then fix. Use when a final image/video completes but looks wrong, such as artifacts, wrong…