BC huawei-cloud-skill-creator
1. Six-phase pipeline for creating Huawei Cloud skills — Socratic requirements gathering, CLI→SDK→API research, MD generation, test preparation, detailed testing, and final cleanup & compliance check 2. Phase-chained dependency: each phase builds on the previous phase's output, no phase may be skipped 3. Supports CLI, SDK, and REST API execution modes with automatic fallback detection 4. Generates complete skill directory structure with SKILL.md, references/, scripts/, templates/ 5. Validates against the Huawei Cloud Skill Specification (华为云Skill检查规范) Triggers include: "创建华为云Skill","新建华为云Skill","华为云skill创建器","创建 Skill","新建 Skill","skill 创建器","create skill","build skill","new skill","skill creator","scaffold a Huawei Cloud skill","wrap CLI or OpenAPI into a skill","package cloud operations into a skill","帮我创建华为云Skill","帮我新建一个Skill","封装华为云CLI为Skill","华为云Skill脚手架","帮我创建一个skill","我需要一个skill","建一个skill","生成skill","帮我建一个华为云skill".
1. Six-phase pipeline for creating Huawei Cloud skills — Socratic requirements gathering, CLI→SDK→API research, MD generation, test preparation, detailed…
As a process C 64/100 · Has gaps — weak spots: result and completion, when it triggers, execution cost
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 2
✓ No critical or high findings
Medium and low: 2
-
medium Dangerous commands
cmd-pipe-to-shellscripts/test-cli-commands.sh:84Downloads and executes remote code from an unrecognised host (pipe to shell) (test fixture / example file)hcloud\ *|python3\ *|curl\ *|bash\ *) return 0 ;;
fixture -
low Risky intent
intent-offensive-securitySKILL.md:324Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)| Vulnerability pattern detection | Known command injection, reverse shell, dangerous function, and prompt injection patterns |
detector
Files scanned: 13. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 8035 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 64/100
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 40Result and completion. Does not say what the result is
- 40Execution cost. Instruction body is 8035 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (bash, web, python, node) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 85Steps. 96 steps, 2 vague phrases
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 12 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (10 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 938: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -212 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 22 example trigger phrases
- +4Structure: 26 headings
- +3Step-by-step instructions: 96 items
- +4Has examples (15 code blocks)
- +4Reference files are cited in the instructions (7 of 7)
- +3All 3 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 69.