AA workflow-guard-rails
Wrap multi-step agent workflows with pre-execution checks, side-effect queues, result validation, retry budgets, checkpointing, audit logs, and failure-rule accumulation. Prevents false successes, duplicate sends, unrecoverable crashes, and silent drift in LLM production systems. Use it when a workflow sends, publishes, pays, deletes, or writes to another system, runs unattended on a schedule, or must be safe to rerun after a mid-task failure. Trigger keywords: workflow safety, workflow guardian, agent guard, guardrails, pre-execution check, pre-flight check, retry budget, idempotency, false success, duplicate send, checkpoint recovery, rerun safety, audit log, drift detection, agent reliability, production guardrails, 工作流守护, Agent 护栏, 副作用队列, 漂移检测, 幂等, 防重复发送, 假成功, 断点恢复, 生产护栏, 重跑安全. 中文摘要：为多步骤 Agent 工作流加装七项护栏——执行前检查、检查点、副作用队列、预算 重试、结果验证、审计记录、规则沉淀，拦截假成功、重复发送与渐进漂移。触发词：工作流 守护、Agent 护栏、假成功拦截、防重复发送、重试预算、断点恢复、生产护栏、漂移检测.
Wrap multi-step agent workflows with pre-execution checks, side-effect queues, result validation, retry budgets, checkpointing, audit logs, and failure-rule…
As a process A 89/100 · Runs to the end — weak spots: running it twice
How to improve
- 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 · 0
✓ No critical or high findings
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "displayName" - note
frontmatter-keyunknown frontmatter key "description_zh" - note
frontmatter-keyunknown frontmatter key "description_en" - note
frontmatter-keyunknown frontmatter key "agent_created" - note
frontmatter-keyunknown frontmatter key "not_for" - note
frontmatter-keyunknown frontmatter key "read_when"
Process rating: all ten parameters 89/100
- 30Running it twice. 8 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Tools and files. No external tools needed
- 100Steps. 28 steps
- 100Result and completion. Output format and completion criterion are stated
- 100When it triggers. States when to use and when not to
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1728 tokens
- 100Progress reporting. Reports progress
- low 13 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)
- +3Description length 925: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +4Structure: 14 headings
- +3Step-by-step instructions: 28 items
- +3Output format is stated explicitly
- +4Has examples (2 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 85.