BC token-guard-pro
Token 守护者是面向 AI Agent 的 token 成本优化系统，针对"压缩过度损失质量、语义缓存命中率低、缺乏模型路由、预算不可见不可控"四大高频痛点而设计。它用三层缓存（精确匹配/语义匹配/模式匹配）+ 自适应压缩 + 模型路由 + 预算守护，在不牺牲响应质量的前提下降低 50-80% 的 token 成本。 核心能力：智能上下文压缩（按重要度分级压缩，代码块与关键决策永不压缩）、三层语义缓存（L1 精确匹配 100% 节省/L2 语义相似 80% 节省/L3 模式匹配 50% 节省）、自适应优化（按 token 压力分阶段调整）、模型路由（按任务复杂度路由到合适模型）、预算守护（设置预算上限，超限告警与降级）、成本可视化（实时报告与趋势分析）、Prefix Cache 支持（重复前缀输入成本降至 1/10）。 适用场景：长会话 token 治理、高频问答场景缓存、多模型混合调用成本优化、企业级 token 预算管控、客服 Agent 成本降低、研究探索类任务的最大化节省。 差异化：相比仅做"上下文压缩"的浅层优化器，本技能新增 (1) 三层语义缓存，L2 语义匹配（cosine 相似度 >= 0.85）显著提升缓存命中率，告别"完全相同才命中"的低效；(2) 模型路由，按任务复杂度自动路由到合适模型（简单任务用小模型，复杂任务用大模型），成本再降 30%；(3) 预算守护，设置日/周/月预算上限，超限自动告警与降级，杜绝预算失控；(4) Prefix Cache 支持，重复前缀输入成本降至 1/10，首 token 延迟降低 50-85%；(5) 成本可视化仪表盘，实时展示节省率、缓存命中率、模型分布、成本趋势。 触发关键词：token优化、成本降低、语义缓存、上下文压缩、模型路由、预算控制、token saver、cost optimization、semantic cache、model routing
Token 守护者是面向 AI Agent 的 token 成本优化系统，针对"压缩过度损失质量、语义缓存命中率低、缺乏模型路由、预算不可见不可控"四大高频痛点而设计。它用三层缓存（精确匹配/语义匹配/模式匹配）+ 自适应压缩 + 模型路由 + 预算守护，在不牺牲响应质量的前提下降低 50-80% 的 token…
As a process C 53/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- 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: 0. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "displayName" - note
frontmatter-keyunknown frontmatter key "summary" - note
frontmatter-keyunknown frontmatter key "tools"
Process rating: all ten parameters 53/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 25 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1847 tokens
- 100Running it twice. No mutating operations
- 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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 828: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +4Structure: 29 headings
- +3Step-by-step instructions: 25 items
- +4Has examples (10 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 72.