DA huawei-cloud-ecs-dsh-deploy
Huawei Cloud Flexus X Instance Deployment - DeepSeek Harness (dsh). ⚠️ Deployment involves actual charges, must confirm region/flavor/billing mode with user before execution. Triggers include: "部署DeepSeek Harness", "部署dsh", "部署DeepSeek智能体", "创建AI智能体服务器", "Deploy DeepSeek Harness", "Deploy dsh", "Create AI Agent Server", "Deploy AI Harness", "DeepSeek Harness", "dsh", "Flexus X", "ECS部署", "AI智能体".
Huawei Cloud Flexus X Instance Deployment - DeepSeek Harness (dsh).
As a process A 82/100 · Runs to the end — no weak spots found
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.
- 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 · 6
-
high Dangerous commands
cmd-install-from-urlreferences/cli-installation-guide.md:39Installs a package from an untrusted URL / archive> pip install -i https://repo.huaweicloud.com/repository/pypi/simple requests huaweicloudsdkcore huaweicloudsdkecs huaweicloudsdkcoc
-
high Dangerous commands
cmd-install-from-urlreferences/cli-installation-guide.md:150Installs a package from an untrusted URL / archivepip install -i https://pypi.tuna.tsinghua.edu.cn/simple huaweicloudsdkcore
-
high Dangerous commands
cmd-install-from-urlreferences/cli-installation-guide.md:153Installs a package from an untrusted URL / archivepip install -i https://repo.huaweicloud.com/repository/pypi/simple huaweicloudsdkcore
Medium and low: 3
-
low Dangerous commands
cmd-background-processscripts/coc_deploy.py:210Starts a background / autostarted process (string literal in code, not executed)systemctl enable dsh
code literal -
low Dangerous commands
cmd-background-processscripts/coc_deploy.py:252Starts a background / autostarted process (string literal in code, not executed)systemctl enable nginx 2>/dev/null || true
code literal -
low Secrets in code
secret-password-literalscripts/deploy_dsh.py:344Hard-coded password / key literal (may be an example) (detector / deny-list definition)args.password = generate_secure_password(12)
detector
Files scanned: 14. 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 ≈ 7645 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "triggers"
Process rating: all ten parameters 82/100
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python, node) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 7645 tokens
- 100Steps. 66 steps
- 100Result and completion. Output format and completion criterion are stated
- 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
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 14 top-level sections: this looks like several domains in one skill
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)
- -4Absolute local paths (C:\Users, /home/…): not portable
- -2localhost URLs: will not work for another user
- -238 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 9 example trigger phrases
- +3Description length 399: enough signal without eating the budget
- +4Structure: 47 headings
- +3Step-by-step instructions: 66 items
- +3Output format is stated explicitly
- +4Has examples (17 code blocks)
- +4Reference files are cited in the instructions (5 of 5)
- +3All 6 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 68.