- Skills
- 25
- 仓库
- 1
- 职业覆盖
- 3
- 场景覆盖
- 4
- Stars
- ★ 0
Bertrand Louargant 的 Skills 25
skill-creator
Create or update a skill playbook (SKILL.md) for the agent. Use when the user wants to create a new skill, add a skill, define a new agent capability, update an existing skill, edit a skill, refine a skill playbook, or specialise the agent for a new domain.
review
Review a body of work (source code, configuration, documents, infra manifests, …) for correctness, clarity, and obvious risks. Use this skill whenever the user asks for a review, audit or critique.
Fallback PDF text extractor using pdftotext (poppler). Use ONLY when the preferred `liteparse` skill is unavailable or for a quick plain-text dump of a simple text PDF — for the best PDF parsing (scanned, multi-column, layout-heavy, or non-PDF formats) prefer the `liteparse` skill.
pandoc
This skill should be used when converting documents between formats (Markdown, DOCX, PDF, HTML, LaTeX, etc.) using pandoc. Use for format conversion, document generation, and preparing markdown for Google Docs or other word processors.
liteparse
Preferred local parser for extracting text from PDFs, images, and scanned/layout-heavy documents (OCR, screenshots, complex PDFs) — far better than pdftotext. Use this FIRST for any PDF, image, or OCR text-extraction or conversion task. NOTE for Office documents (DOCX, PPTX, XLSX, ODT) — for plain text extraction prefer the `pandoc` skill, which needs no LibreOffice; only use LiteParse for Office docs when you need its layout/OCR/screenshot output. The `pdf` skill (pdftotext) is only a fallback for when LiteParse cannot be installed.
linux-admin
Safely make a change to a Linux workstation — install/remove packages, manage systemd services and timers, edit and validate configuration files, manage users/permissions and networking. Detect the distro and package manager, determine whether the target is package-managed or hand-maintained, preview or simulate every change, back up configs before editing, and prefer the package manager and drop-in overrides so the host stays consistent and secure. Use whenever a change to the local Linux host is required (install, remove, enable, disable, start, stop, edit a config, add a user, change permissions or networking).
k8s-triage
Diagnose an unhealthy Kubernetes workload — pods crash-looping, deployments not ready, services not reachable. The triage decision playbook — confirm context, get evidence, classify the failure, propose one safe next action. Use whenever the user mentions kubernetes, k8s, kubectl, pods, deployments, namespaces, or attaches a kubectl error.
k8s-modification
Safely modify a live Kubernetes resource — detect whether it is Helm-, GitOps- (Flux/Argo), or kubectl-managed; preview every change with kubectl/helm diff and dry-run; prefer Helm-native changes for Helm releases; and fall back to server-side apply / strategic-merge patch with care so future helm upgrades are not broken. Use whenever a change to a live cluster is required (apply, patch, scale, set image, edit values, helm upgrade).
k8s-log-investigation
Investigate Kubernetes pod logs efficiently when logs are large. Use whenever the user mentions kubernetes, k8s, pod logs, errors, warnings, exceptions, crash loops, or asks to diagnose logs without dumping everything.
k8s-investigation
Gather read-only evidence about an unhealthy Kubernetes workload — the snapshot mechanics (get/describe/logs/events) that back a triage. Use whenever you need to collect cited kubectl evidence for a pod/deployment/service before diagnosing it, without mutating the cluster.
k8s-cleanup
Remove ephemeral Kubernetes diagnostic leftovers the squad created — debug pods, throwaway run pods, one-off jobs, temporary configmaps/secrets — found by the omnis.dev/ephemeral=true label (well-known debug-pod name patterns are surfaced as confirm-first suspects). Removes only ephemeral resources, never real workloads. Use to sweep leftovers at the end of an investigation or on request.
k8s-audit
Audit a Kubernetes cluster for policy compliance — enumerate every resource in scope and report the EXACT set that violates a stated rule (e.g. missing resource limits/probes/labels, disallowed image repos or tags, RBAC bindings, PodDisruptionBudgets, admission/OPA-Gatekeeper/Kyverno constraints). Use whenever the task is "list/find all resources that violate policy X" or a security-posture / compliance check across many resources, as opposed to diagnosing one unhealthy workload.
deep-research
The playbook for an EXHAUSTIVE, decision-grade research investigation — when the user asks for deep/in-depth/thorough/exhaustive research, says they need it to make a decision, or asks whether an approach/tool/configuration is a good idea. Audits the assumptions baked into the question, runs multiple search waves with a coverage review between them, considers the alternatives the user did not mention, and has the findings adversarially reviewed before answering. Use instead of a single search-and-summarise pass whenever being incomplete would cost the user a bad decision.
ask-user
Pattern for asking the user structured questions and gating execution on their response. Use whenever the agent needs a choice, confirmation, or free-text answer before it can proceed. Renders an interactive widget in the web UI and TUI, and falls back to stdin on the console.
agent-builder
Scaffold a new specialised agent and wire it into the harness. Use when the user says "create an agent", "add a specialist", "build a new agent", or wants to extend the harness for a new domain. Covers tool composition, skill authoring, MCP server wiring, permissions, and coordinator registration.
demo-skill
A demo skill shipped by the demo plugin.
web-docs-update
Audit and update the web UI documentation (web/docs/*.md) after changes to the web UI source files (web/app.js, web/settings.js, web/index.html, server/*.go). Use when the user modified the web UI and wants the docs kept in sync, or when a doc review reveals stale content.
root-wiring
Understand and modify the wiring of main.go (root) — how tools, sub-agents, plugins and toolsets are assembled into the lead agent. Use when adding a sub-agent, swapping a plugin, changing the toolset list, or debugging why a feature isn't reaching the agent. Mention triggers - root binary, main.go, wiring, lead agent, agenttool, NewMultiLoader, plugins.