BB huawei-cloud-skill-tester
End-to-end functional testing framework for Huawei Cloud skills — three-tier pipeline covering single-skill unit testing, multi-skill orchestration, and end-to-end full flow testing. Each phase produces structured JSON output with chain verification. Supports skill installation validation, functional analysis, CLI→SDK→API feasibility research, test case generation, real-environment execution with resource lifecycle, resource cleanup, multi-skill scenario orchestration, trigger-conflict detection, and consolidated reporting. Triggers include: 测试技能, 执行技能测试, 跑测试流程, 技能回归测试, skill test, run skill tests, test huawei cloud skill, verify skill, 测试华为云skill, 全流程测试, 编排测试, 技能完整性检查, skill-tester, 跑测试, 回归测试, 组合测试, 多skill编排, verification, e2e.
End-to-end functional testing framework for Huawei Cloud skills — three-tier pipeline covering single-skill unit testing, multi-skill orchestration, and…
As a process B 66/100 · Nearly there — weak spots: when it triggers, running it twice, progress reporting
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
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
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- 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 · 9
-
high Dangerous commands
cmd-pipe-to-shellreferences/cli-installation-guide.md:10Downloads and executes remote code from an unrecognised host (pipe to shell)curl -sSL https://apiexplorer.developer.huaweicloud.com/install/hcloud/install.sh | bash
Medium and low: 8
-
medium Dangerous commands
cmd-eval-dynamicscripts/tier2/phase-6-full-flow.sh:108Dynamic code execution from decoded/untrusted inputexec(open(_PU, encoding='utf-8').read())
-
low Secrets in code
secret-high-entropy-tokenreferences/agent-protocol.md:17High-entropy token-like string (may be an id, hash or a credential)__HU…1__
-
low Secrets in code
secret-high-entropy-tokenreferences/agent-protocol.md:62High-entropy token-like string (may be an id, hash or a credential)[Phase 4] __HU…1__
-
low Dangerous commands
cmd-eval-dynamicscripts/lib/placeholder-utils.py:6Dynamic code execution from decoded/untrusted input (code comment)# exec(open("<SCRIPT_DIR>/lib/placeholder-utils.py", encoding='utf-8').read())comment -
low Secrets in code
secret-high-entropy-tokenscripts/lib/utils.sh:302High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)CRED_REQUEST_SENTINEL="__HU…1__"
quoted -
low Dangerous commands
cmd-eval-dynamicscripts/tier1/phase-3-gen-testcases.sh:37Dynamic code execution from decoded/untrusted input (test fixture / example file)exec(open(os.environ.get('PLACEHOLDER_UTILS', '')).read())fixture -
low Dangerous commands
cmd-eval-dynamicscripts/tier1/phase-4-execute-tests.sh:74Dynamic code execution from decoded/untrusted input (test fixture / example file)exec(open(os.environ.get('PLACEHOLDER_UTILS', '')).read())fixture -
low Secrets in code
secret-high-entropy-tokenSKILL.md:366High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)When Phase 4 or Phase 6 needs to call live Huawei Cloud APIs but cannot find credentials in the environment, the framework does **not** silently skip. It emits a structured request (sentinel line `__H
quoted
Files scanned: 28. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5471 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 66/100
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 9 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 5471 tokens
- 85Steps. 67 steps, 1 vague phrases
- 100Result and completion. Output format and completion criterion are stated
- 100Consistency. Name and required fields are in place
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 15 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (13 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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +1No license
- +2Single-language instructions
- +3Description length 738: enough signal without eating the budget
- +4Structure: 28 headings
- +3Step-by-step instructions: 67 items
- +3Output format is stated explicitly
- +4Has examples (8 code blocks)
- +4Reference files are cited in the instructions (9 of 9)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.