DB li_maestro_evaluate
Interactive Q&A Threat Modeling — conversational CSA MAESTRO risk assessment for agentic AI systems AND OpenCode Skills (.md/.docx/.xlsx), with AI risk classification mapping to 《人工智能安全治理框架》2.0. Supports two analysis modes: MVTM Checklist (minimum viable threat model with Chinese regulatory extensions) and Full 10-phase assessment. | 基于CSA MAESTRO框架的交互式问答威胁建模评估，面向智能体AI系统及OpenCode Skill，输出多格式(.md/.docx/.xlsx)风险评估报告及AI安全风险分类对照表。支持MVTM检查表与全面评估两种模式。
Interactive Q&A Threat Modeling — conversational CSA MAESTRO risk assessment for agentic AI systems AND OpenCode Skills (.md/.docx/.xlsx), with AI risk…
As a process B 67/100 · Nearly there — weak spots: when it triggers, consistency, execution cost
How to improve
- 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 · 5
✓ No critical or high findings
Medium and low: 5
-
low Risky intent
intent-offensive-securitySKILL.md:1777Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Can multiple installed skills produce harmful emergent behavior when activated together? | SK-C1 | A2A Communication, Autonomy | Individual skills pass review but their combined prompt+tool access c
A further 4 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 9. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
name-formatname should be kebab-case (lowercase letters, digits, hyphens) - warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 35602 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 67/100
- 10Execution cost. Instruction body is 35602 tokens: crowds the task out of the window
- 20When it triggers. No condition that starts the skill
- 40Consistency. Frontmatter name (li_maestro_evaluate) differs from the folder (li-maestro-evaluate)
- 60Tools and files. Uses tools (bash, web, python, node) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 85Steps. 123 steps, 3 vague phrases
- 100Result and completion. Output format and completion criterion are stated
- 100Failures and branches. 7 branches, has a failure section
- 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 35 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (7 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)
- -5TODO / placeholder text left in the skill
- -217 emoji in the instructions: noise for the model
- -31 of 3 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +3Description length 449: enough signal without eating the budget
- +4Structure: 138 headings
- +3Step-by-step instructions: 123 items
- +3Output format is stated explicitly
- +4Has examples (57 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 51.