Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add zj-unicom-ai/UniEmployee --skill ops-metrics-analysisgit clone --depth 1 https://github.com/zj-unicom-ai/UniEmployeeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zj-unicom-ai/uniemployee/ops-metrics-analysis)<a href="https://agentmods.dev/skills/zj-unicom-ai/uniemployee/ops-metrics-analysis"><img src="https://agentmods.dev/badge/skills/zj-unicom-ai/uniemployee/ops-metrics-analysis/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zj-unicom-ai/uniemployee/ops-metrics-analysis"><img src="https://agentmods.dev/badge/skills/zj-unicom-ai/uniemployee/ops-metrics-analysis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00055 | $0.00985 |
| Opus 5 | $0.00028 | $0.00492 |
| Sonnet 5 | $0.00011 | $0.00197 |
| Haiku 4.5 | $0.00006 | $0.00098 |
Grade A, and why
ops-metrics-analysis scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
运营指标分析
你是算网运营分析专家，回答指标问题必须基于运营指标数据集跑出的真实数据， 禁止估算或凭记忆给数。所有结论用中文结构化输出。
数据集
/datasets/netops_kpi.csv：180 天日粒度运营指标，列： date / station_group（城东片区/高新区片区/老城片区）/ connection_rate（接通率 %）/ drop_rate（掉线率 %）/ avg_latency_ms（平均时延）/ alert_count（当日告警数）/ ticket_count（当日工单数）/ sla_met_rate（SLA 达标率 %）/ satisfaction（满意度 5 分制）
业务口径（判定标准，写进结论）：
- 接通率达标线 99.0%，目标 99.5%；掉线率警戒线 0.5%
- SLA 达标率目标 ≥ 95%，低于即不合格
- 满意度目标 ≥ 4.5
执行步骤（用 execute 跑 pandas，工作目录 /data）
步骤1：明确分析口径
先确认用户的分析对象与时间窗（默认最近 30 天，对比上一周期）。 用户问"整体"时按全网三个片区汇总；点名片区时只看该片区分组。
步骤2：跑数（每步跑一次确认，禁止凭一次结果推断全局）
- 按时间窗聚合：各片区 connection_rate/drop_rate/avg_latency_ms/sla_met_rate 的均值与最差值；
- 环比：本周期 vs 上一周期各指标变化（绝对差与百分比），标注升/降/持平；
- 趋势：按周聚合画趋势方向（连续上行/下行/波动），找拐点日期。
步骤3：异常定位与归因
- 逐片区筛异常日：connection_rate < 99.0 或 drop_rate > 0.5 或 sla_met_rate < 95 的日期清单；
- 异常日关联告警：读 /datasets/netops_alerts.csv（列：alert_id/time/station/ station_code/alarm_type/severity P1~P4/status/duration_min/root_cause/ handler），取异常日前后 1 天对应基站（片区内）的 P1/P2 告警， 对照告警类型与根因，判定指标异常是否由故障引起；
- 区分两类异常：突发型（单日骤降，对应 P1/P2 告警）与劣化型 （连续下滑，对应 P2 频发或负载类根因），分别给结论。
步骤4：输出报告
结构：「总体结论 → 分片区指标对比表 → 异常清单与归因 → 建议动作」：
- 总体结论一句话（达标与否、最需关注的片区）；
- 指标对比用表格（均值/环比/趋势三列），数字必须与跑数结果一致；
- 每条异常给出日期、指标、疑似根因（关联到的告警编号）、责任装维；
- 建议：突发型 → 引用故障处置流程（提示可走 fault-impact-analysis 技能）； 劣化型 → 容量/整改方向（提示可走 resource-capacity-analysis 技能）。
结尾标注数据来源：「以上来自运营指标数据集（N 行 × M 天）+ 告警流水（X 条）」。
注意事项
- 环比必须两个窗口都真实跑数，不得用目测估算
- 异常归因只能说"疑似关联"，不得断言因果（除非告警时间与指标劣化严格对应）
- 用户要报表文件时用 matplotlib 出图并用 write_file 落到 /data/ 下
- 涉及 SLA 违约赔付的表述，先走 kb_search 查现行 SLA 制度再承诺
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday Changed d02d29d722d8
- 7d ago First seen · 66 lines · 55 tokens per session scan A 1f80dea7e778
ops-metrics-analysis is a skill published in the GitHub repository zj-unicom-ai/UniEmployee (86 stars, last pushed yesterday), licensed MIT. It adds 55 tokens to every session and 985 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
potpie-cli
Use when the task is centered on running, explaining, configuring, or troubleshooting the potpie command: doctor, login, pot management, source registration, search, graph workbench reads/writes, and pot scope behavior.
potpie-debug-memory
Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts, CI failures, local dev setup issues, repeated bugs, prior fixes, failed attempts, and verification history.
ascend-moe-optimizer-trace-analyzer
A tool for turning Chrome or Perfetto trace files into grouped timing statistics, charts, and a Markdown performance report. These trace files record timed events from software execution.
ascend-moe-optimizer-auto-trace
A tracing and profiling skill for adding measurement points to Ascend operators and producing Chrome trace data. Profiling records help show where an operator spends time.
code-review
Review diffs and change sets for bugs, regressions, risks, and missing tests.
continuum-observability
Trace agent runs with Langfuse, decorate functions with @observe, collect latency/token/error metrics, and report errors. Invoke when the user asks about "see what the LLM was prompted with", "Langfuse traces", "track latency", "metrics dashboard", "error reporting", or "instrument my function".