Borrowing it
Nothing to install: this file belongs to victoriacity/openakari. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/victoriacity/openakari/main/.claude/skills/diagnose/SKILL.mdgit clone --depth 1 https://github.com/victoriacity/openakariWrote 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/victoriacity/openakari/diagnose)<a href="https://agentmods.dev/skills/victoriacity/openakari/diagnose"><img src="https://agentmods.dev/badge/skills/victoriacity/openakari/diagnose.svg" alt="Measured on agentmods" 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 | $0.00015 | $0.01846 |
| Opus 5 | $0.00008 | $0.00923 |
| Sonnet 5 | $0.00003 | $0.00369 |
| Haiku 4.5 | $0.00002 | $0.00185 |
Grade A, and why
diagnose 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 8d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/diagnose
You are diagnosing empirical results — finding patterns in errors, generating hypotheses about root causes, and assessing whether the results mean what they appear to mean. This is the analytical complement to /synthesize (which works across accumulated findings); /diagnose works within one result set.
The argument is a path to results (CSV, log entry, analysis output) or a description of what to examine. Read the data first.
When to use this vs alternatives
- Use
/diagnosewhen you have empirical results (CSVs, metrics, error logs) and want to understand what they mean — error patterns, root-cause hypotheses, validity assessment. - Use
/postmortemwhen the problem is not "what do the results mean?" but "why did an agent report flawed results as correct?" Postmortem analyzes reasoning failures; diagnose analyzes data. - Use
/review metricswhen you suspect the metrics themselves may be degenerate or misleading before interpreting the results./reviewchecks whether results are interpretable;/diagnoseinterprets them.
Step 1: Understand the experiment
- Read the results file and any associated project log entries.
- Identify what was measured, what was varied, and what the expected outcome was.
- Read the experiment design or method description if one exists.
- Identify which CI layers are involved.
Step 2: Characterize the error distribution
Do not start with individual examples. Start with the distribution:
- Overall rates: What is the base rate of success/failure? How does it compare to random chance or a naive baseline?
- Conditional rates: Break errors down by every available dimension (model, question type, category, run number, etc.). Where are errors concentrated?
- Error types: Categorize errors. Are they systematic (same direction, same condition) or random (scattered)? Common categories for judgment tasks:
- Wrong direction (picked A when B was correct)
- False consensus (called a tie when humans disagreed)
- False distinction (picked a winner when humans saw a tie)
- Magnitude error (correct direction but wrong confidence)
- Temporal patterns: Do error rates change over runs? Is there a position effect, order effect, or fatigue analog?
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.
- 8d ago First seen · 159 lines · 15 tokens per session scan A 5fdac93b826f
diagnose is a skill published in the GitHub repository victoriacity/openakari (47 stars, last pushed 6mo ago), licensed MIT. It adds 15 tokens to every session and 1,846 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
trace
Evidence-driven tracing lane that orchestrates competing tracer hypotheses in Claude built-in team mode.
smiles-validation
Strict SMILES validation, structural comparison, and modification verification. Catches invalid LLM-generated molecules.
cnsplots
Create, revise, and troubleshoot publication-ready scientific plots in Python with cnsplots, including distribution, regression, heatmap, genomics, survival, set, flow, and multi-panel figures. Use when a user asks for cnsplots code, Cell/Nature/Science-style visualization, precise physical figure dimensions…
social-physics-filter
Unified boundary enforcement, interpersonal diagnostic, and relational audit engine. Absorbs 40 psychology + 2 social protocols and all relationship case studies.
relax-dev-debug
Develop and debug the Relax reinforcement learning project. Use this skill whenever modifying code in the relax/ directory, or running remote training jobs on a Ray cluster for validation. Also use it when the user mentions training, debugging training runs, submitting Ray jobs, or fixing training errors.
thinking-scientific-method
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.