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 huaweicloud/huaweicloud-skills --skill huawei-cloud-cdn-abnormal-status-code-analysisgit clone --depth 1 https://github.com/huaweicloud/huaweicloud-skillsWrote 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/huaweicloud/huaweicloud-skills/huawei-cloud-cdn-abnormal-status-code-analysis)<a href="https://agentmods.dev/skills/huaweicloud/huaweicloud-skills/huawei-cloud-cdn-abnormal-status-code-analysis"><img src="https://agentmods.dev/badge/skills/huaweicloud/huaweicloud-skills/huawei-cloud-cdn-abnormal-status-code-analysis/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/huaweicloud/huaweicloud-skills/huawei-cloud-cdn-abnormal-status-code-analysis"><img src="https://agentmods.dev/badge/skills/huaweicloud/huaweicloud-skills/huawei-cloud-cdn-abnormal-status-code-analysis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 118 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00231 | $0.04018 |
| Opus 5 | $0.00115 | $0.02009 |
| Sonnet 5 | $0.00046 | $0.00804 |
| Haiku 4.5 | $0.00023 | $0.00402 |
Grade A, and why
huawei-cloud-cdn-abnormal-status-code-analysis 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 — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CDN Abnormal Status Code Analysis
Overview
This skill diagnoses abnormal HTTP status codes (4xx / 5xx) on a Huawei Cloud CDN accelerated domain. It is read-only end to end: every action is a query (GET statistics / GET config / GET log-download-link / read-only log fetch). It answers four questions for any 4xx/5xx anomaly:
- Is there an anomaly, and how big? — quantify 4xx/5xx volume and ratio over a window.
- Which exact code, when, edge or origin? — drill to the specific code + time bucket, and fork edge-generated vs origin-generated via the back-to-source (
bs_*) status statistics. - Who / where / what path? — top IPs/paths/UA/referer, client IP count (刷量 vs real users).
- Root cause + forensics? — narrow to edge-config rules or origin-side issues, then pin the exact requests via access logs.
Key Features:
- Edge-vs-origin fork driven by one command (
bs_status_code_*):result={}⇒ edge-generated; non-empty ⇒ origin-generated. - Covers the full 4xx/5xx set: 400/401/403/404/405/406/408/410/412/416/429/499/495/496/500/501/502/503/504/505/530….
- Read-only log helper
scripts/fetch_cdn_log.pydownloads + decompresses a CDN log and extracts the abnormal-status rows as JSON. - Structured diagnosis report with a status-code → root-cause decision tree and console/工单 remediation boundary (no CLI write ops).
Tools: hcloud CLI (KooCLI) + Python log helper (scripts/fetch_cdn_log.py).
Core Principle: Read-only diagnosis; no configuration changes.
⛔ Prohibited Operations (Security Constraints)
This skill strictly prohibits all non-GET (write/modify/delete) CDN operations, even if requested by the user.
Total: 55 prohibited operations (24 POST + 25 PUT + 6 DELETE).
For the complete list of all 55 prohibited non-GET operations with risk descriptions, see references/prohibited-operations.md.
Representative prohibited operations (full list in the reference doc):
What ships with it
16 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.
- .markdownlint.json 359 B
- references/acceptance-criteria.md 3.6 KB
- references/cli-installation-guide.md 1.7 KB
- references/dataflow-diagram.md 3.5 KB
- references/iam-policies.md 2.4 KB
- references/prohibited-operations.md 6.4 KB
- references/related-apis.md 7.1 KB
- references/task-discovery.md 2.3 KB
- references/task-distribution.md 1.5 KB
- references/task-forensics.md 3.9 KB
- references/task-localize.md 2.1 KB
- references/task-report.md 2.9 KB
- references/task-rootcause.md 4.3 KB
- references/troubleshooting.md 2.9 KB
- references/verification-method.md 2.3 KB
- scripts/fetch_cdn_log.py 8.7 KB runs code
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 · 227 lines · 231 tokens per session scan A 5fcc1c0a1d78
huawei-cloud-cdn-abnormal-status-code-analysis is a skill published in the GitHub repository huaweicloud/huaweicloud-skills (49 stars, last pushed yesterday), licensed MIT. It adds 231 tokens to every session and 4,018 once invoked, about $0.0012 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 2026-09-03.
Other skills, from other repositories
diagnose-electrical-fault
Use when a vehicle has a blown fuse, inoperative accessory, warning light, or intermittent electrical failure — applying a systematic test sequence (visual, OBD-II codes, voltage/continuity testing) to isolate the fault to a specific component or circuit without replacing parts blindly.
troubleshooting
A troubleshooting guide for Kubernetes, a system for running containerized applications across computers. It lists commands for checking clusters, nodes, components, pods, events, resources, and storage.
debugging-strategies
Systematic debugging methodologies for diagnosing and resolving issues.
avatar-troubleshoot
A troubleshooting guide for finding and diagnosing problems in virtual-person integrations, such as digital avatars connected to other software.
Debugger
Systematic issue diagnosis skill that isolates root causes, tests hypotheses, and produces minimal fix strategies.
trulens-diagnosis
Diagnose low evaluation scores and generate actionable improvement recommendations.