FB huawei-cloud-ascend-remote-connect
Provides temporary SSH remote connection for Huawei Cloud Ascend devices with dynamic host/port/user/password input (in-memory only), disk management, NPU monitoring, container management, security auditing and log analysis; sensitive operations require user confirmation before execution Use this skillthis skill to remotely connect to Ascend servers, monitor NPU health, manage disks/LVM, or troubleshoot Ascend devices. Triggerer words: SSH, remote, Ascend, NPU, 远程连接, 昇腾, NPU监控
Provides temporary SSH remote connection for Huawei Cloud Ascend devices with dynamic host/port/user/password input (in-memory only), disk management, NPU…
As a process B 77/100 · Nearly there — weak spots: when it triggers
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 instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
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.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
How to improve
- Remove the critical guard findings (secrets, dangerous commands, hidden instructions): while they stand the skill is blocked and cannot grade above F.
- 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
-
critical Exfiltration
exfil-post-credential-filescripts/executor.py:443Uploads a credential/secret file to a remote endpointself.pending_confirmation = 'ssh-keygen -t ed25519 -f /root/.ssh/id_ed25519 -N "" && cat /root/.ssh/id_e…pub'
Medium and low: 5
-
medium Dangerous commands
cmd-destructive-fs-scopedscripts/executor.py:941Destructive command on a partition / current directory (normal in admin docs; verify the target) (string literal in code, not executed)mkfs.ext4 /dev/{vg_name}/{lv_name}code literal -
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: bashallowed-tools: python3 bash ssh
-
low Dangerous commands
cmd-cron-mentionscripts/command_validator.py:51Mentions editing / listing crontab (string literal in code, not executed)'crontab -l', 'passwd', 'su', 'sudo', 'apt', 'apt-get', 'yum',
code literal -
low Dangerous commands
cmd-cron-mentionscripts/executor.py:249Mentions editing / listing crontab (string literal in code, not executed)return self._exec_simple('scheduled tasks', 'echo "=== root crontab ==="; crontab -l 2>/dev/null || echo "none"; echo; for user in $(cut -d: -f1 /etc/passwd | head -10); do crontab -u $user -l 2>/dev/code literal -
low Dangerous commands
cmd-destructive-fsSKILL.md:436Destructive filesystem command (wipes root/home/drive) (negated — the text forbids it)- `:(){ :|:& };:` (fork bomb)negated
Files scanned: 11. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 77/100
- 20When it triggers. No condition that starts the skill
- 60Result and completion. Output format stated, no completion criterion
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 4077 tokens
- 85Steps. 122 steps, 1 vague phrases
- 100Tools and files. Tools declared in frontmatter
- 100Failures and branches. 2 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 20 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
- +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 481: enough signal without eating the budget
- +4Structure: 67 headings
- +3Step-by-step instructions: 122 items
- +3Output format is stated explicitly
- +4Has examples (18 code blocks)
- +4Reference files are cited in the instructions (3 of 3)
- +3All 6 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 94.