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 agentmods add skills/hugorcd/evlog/linear-maintenancenpx skills add HugoRCD/evlog --skill linear-maintenancegit clone --depth 1 https://github.com/HugoRCD/evlogWhat 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 | $0.00089 | $0.01054 |
| Opus 5 | $0.00044 | $0.00527 |
| Sonnet 5 | $0.00018 | $0.00211 |
| Haiku 4.5 | $0.00009 | $0.00105 |
Grade A, and why
linear-maintenance 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 3d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Linear maintenance
A regular housekeeping sweep over the evlog team's Linear backlog so it stays readable and current between passes. The goal is a tidy, trustworthy backlog, not a census. Listing every issue is not the point; correcting the drift and flagging what needs a decision is.
Where the other skills draw the line: self-review owns the coherence and
reach of the agent's own surface, repo-health-sweep owns the repository and
docs, cost-watchdog owns gateway spend. This skill owns Linear housekeeping
only, and stops there. When a pass finds evidence that shipped work is now
live (a schedule, a skill, a merged PR), the finding belongs to this skill's
"close it" bucket, not to building anything.
Two halves, in order
Apply what is mechanically verifiable. These are the unambiguous fixes,
and they are the point of the run: the sibling set or the issue history is
evidence enough that a write is correct. Apply them directly via
linear__save_issue.
Surface what needs a judgement call about intent. Anything you cannot verify, or whose correct value is a matter of taste, stays open and is written down as a finding for Hugo. Never guess a decision that a sibling set or the issue history cannot support.
The lenses
Enumerate the open surface first: linear__list_issues on the evlog team,
across states. Then for each item check:
- Project: every open issue sits in the project its title and content
point to, or in
Triageif it is genuinely unplaced. An issue whose peers carry a project but it does not is a fix, not a finding. - Priority: an item
In Progressshould carry a priority consistent with its siblings; aBacklogitem with no priority next to prioritized peers is inconsistent. A priority change that relies on an estimate of intent is a finding. - Label: every open issue carries one of the team's labels. A feature request with no label is a fix. A label that duplicates another, or has a single use, is a hygiene finding.
- Stale: triage older than several months with no movement, parked next to a canceled sibling, or an investigation that outlived its question, is a finding (close it or kick it to the community).
- Done but open: an item whose work already shipped in another issue, a schedule, or a merged PR, and is still open, is a close. Only when the shipped evidence is verifiable from the issue history or linked work.
- Consistency: a true duplicate of a completed or archived sibling, a sub-issue misread as a duplicate, work whose state contradicts its neighbors.
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.
- 3d ago First seen · 85 lines · 89 tokens per session scan A 677b88dc6648
linear-maintenance is a skill published in the GitHub repository HugoRCD/evlog (1,831 stars, last pushed yesterday), licensed MIT. It adds 89 tokens to every session and 1,054 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
frontmcp-observability
Use when adding tracing, structured logging, metrics, or monitoring to a FrontMCP server. Covers zero-config OpenTelemetry distributed tracing across all flows; the this.telemetry API for custom spans, events, and attributes in tools, plugins, agents, and skills; structured JSON logging with trace correlation and…
golang-observability-opentelemetry
Instrumenting Go applications with OpenTelemetry for distributed tracing, Prometheus for metrics, and structured logging with slog.
Observability Checklist
Reviews a service or codebase against a full observability checklist — logs, metrics, traces, and alerting gaps.
anyrobot-shadcn-frontend-js
AnyRobot 前端开发核心指南：提供 JavaScript + React + shadcn/ui 开发的精简实用规范和最佳实践。 当用户需要前端开发指导、组件开发或项目结构建议时触发。 本技能提供实际可用于项目开发的核心代码示例和开发指南。.
数据模型文档生成
当用户要生成或更新 Doris 表的数据模型 Markdown（三一级标题：数据摘要、数据表列、JSON扩展字段）、从 Doris 拉取元数据与至少 100 行样本，并由脚本或 --llm 补全说明时启用。.
fix-loop
Use when a Java app that has stacktale installed is failing — after running it or its tests, to find what broke, fix it, and confirm the fix by re-running and checking for new errors. Also use when asked to "check the errors", "what's failing", or to work through errors-ai.log.