BB alibabacloud-pai-eas-service-diagnose
PAI-EAS service diagnosis and troubleshooting. Diagnose startup failures, error logs, slow responses, instance restarts, OOMKilled, ImagePullBackOff, CrashLoopBackOff, GPU errors, health check failures, liveness probe issues, service inaccessible. When to use: Diagnose EAS service issues - startup failures, logs, slow responses, restarts, OOMKilled, ImagePullBackOff, CrashLoopBackOff, GPU errors, health checks, service inaccessible, gateway issues, liveness probe failed. Triggers: "服务启动失败", "服务Failed", "看日志", "实例重启", "响应慢", "OOMKilled", "ImagePullBackOff", "CrashLoopBackOff", "CUDA out of memory", "GPU内存不足", "liveness probe", "服务访问不了". Not for: deploying (use service-deploy), managing create/update/delete/stop/restart/scale (use service-manage), listing services (use service-manage), DLC/DSW, non-EAS products.
As a process B 69/100 · Nearly there — weak spots: result and completion, inputs and preconditions
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.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
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.
How to improve
- 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
✓ No critical or high findings
Medium and low: 6
-
medium Dangerous commands
cmd-privilegereferences/cli-installation-guide.md:43Privilege escalation / world-writable permissionssudo mv aliyun /usr/local/bin/
-
medium Dangerous commands
cmd-privilegereferences/cli-installation-guide.md:58Privilege escalation / world-writable permissionssudo mv aliyun /usr/local/bin/
-
medium Dangerous commands
cmd-privilegereferences/cli-installation-guide.md:71Privilege escalation / world-writable permissionssudo mv aliyun /usr/local/bin/
-
medium Dangerous commands
cmd-privilegereferences/cli-installation-guide.md:84Privilege escalation / world-writable permissionssudo mv aliyun /usr/local/bin/
-
medium Exfiltration
net-credential-usereferences/diagnosis-flow.md:194Credential used in a network call (verify the destination is the intended service)curl --connect-timeout 10 --max-time 30 -H "Authorization: $TOKEN" "$ENDPOINT/health"
-
low Dangerous commands
cmd-privilegereferences/cli-installation-guide.md:9Privilege escalation / world-writable permissions (quoted — discussed, not commanded)> **Note on `sudo mv ... /usr/local/bin/`**: The binary-install steps below use
quoted
Files scanned: 11. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5666 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 69/100
- 30Inputs and preconditions. Does not say what the process needs to start
- 40Result and completion. Does not say what the result is
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 70Execution cost. Instruction body is 5666 tokens
- 100Steps. 25 steps
- 100Failures and branches. 19 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. No mutating operations
- 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
- low The response is described with custom markup (16 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
- +3Description length 823: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 9 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 20 headings
- +3Step-by-step instructions: 25 items
- +4Has examples (14 code blocks)
- +4Reference files are cited in the instructions (9 of 9)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 85.