BB alibabacloud-agent-identity-agentrun-e2e
Stand up a working Alibaba Cloud agent that authenticates its end users and proves its permissions hold, then tear it back down. Use when someone wants to deploy an AgentRun-hosted agent governed by AgentIdentity, see for themselves that only signed-in users reach it, that each tool receives the credentials it needs without any secret in the code, and that authorization rules actually block what they should — or when they want to remove the resources such a trial created. Covers first-time setup on an empty account, re-verification after a change, and cleanup. 也响应"AgentRun 端到端测试""AgentIdentity 全流程验证""AgentRun 部署 Agent"等中文请求。
Stand up a working Alibaba Cloud agent that authenticates its end users and proves its permissions hold, then tear it back down.
As a process B 70/100 · 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 skill tells the agent to hide things from you: not to show errors, not to mention actions, to report differently from what was done. You lose the ability to see what the agent really did.
Transparency beats a smooth answer. If the goal is to hide technical noise, ask the agent to "summarise briefly", not to "not mention".
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 · 2
-
high Concealment
en-hide-from-userreferences/console-guides.md:87Instruction to hide actions from the userassume one exists and never tell the user what their account "already has";
Medium and low: 1
-
low Secrets in code
secret-password-literalscripts/01_prepare_identities.sh:100Hard-coded password / key literal (may be an example)apikey="demo…123"))
Files scanned: 18. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5429 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 70/100
- 20When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 5429 tokens
- 85Steps. 38 steps, 1 vague phrases
- 100Failures and branches. 1 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 13 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (11 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)
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +3Description length 632: enough signal without eating the budget
- +4Structure: 14 headings
- +3Step-by-step instructions: 38 items
- +3Output format is stated explicitly
- +4Has examples (15 code blocks)
- +4Reference files are cited in the instructions (7 of 7)
- +3All 8 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 90.