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 holoviz-dev/holoviz-skills --skill cleanupgit clone --depth 1 https://github.com/holoviz-dev/holoviz-skillsWrote 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/holoviz-dev/holoviz-skills/cleanup)<a href="https://agentmods.dev/skills/holoviz-dev/holoviz-skills/cleanup"><img src="https://agentmods.dev/badge/skills/holoviz-dev/holoviz-skills/cleanup/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/holoviz-dev/holoviz-skills/cleanup"><img src="https://agentmods.dev/badge/skills/holoviz-dev/holoviz-skills/cleanup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00038 | $0.01152 |
| Opus 5 | $0.00019 | $0.00576 |
| Sonnet 5 | $0.00008 | $0.00230 |
| Haiku 4.5 | $0.00004 | $0.00115 |
Grade A, and why
cleanup 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 10d 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cleanup
This skill covers code quality patterns and common pitfalls when reviewing or refactoring HoloViz code.
Contents
Review
- Perform a
git difffrom the PR branch to the main branch and review for potential issues, improvements, and adherence to best practices. - Consider the full set of changes and whether there is a simpler way to achieve the same result. A PR that touches five files to work around a problem may have a two-line fix elsewhere.
- Explain why this approach over the alternatives (mixin vs. inheritance vs. duplication); reviewers consistently ask for that rationale.
- Reuse the naming and option behavior of similar existing components instead of inventing a new spelling for the same idea (e.g.
font_sizeto match a sibling element, notfontsize). - Don't change an existing default or signature — that breaks users — unless a breaking change is the explicit goal of the PR.
- Share cross-backend/variant logic via a mixin or helper, but extract only what is genuinely common.
- Scrutinize AI-assisted code like any other; flag it and verify the behavior yourself.
Code Style
- Leave formatting and style enforcement (including type hint syntax) to linters and pre-commit hooks. Run via
pixi run lint. - Top-level imports should only be from the standard library, required dependencies, and relative imports. Imports of optional or slow-loading dependencies should go inside the function that uses them.
- Prefer direct attribute access when the attribute is known to exist.
getattrwith a default is appropriate when the attribute may be absent (e.g. checking across class hierarchies or optional mixins) and the caller handles the fallback. - Order file contents: imports, constants, functions (or a
utilsmodule), then classes. @staticmethodis fine when the method is part of the class's public interface or is only meaningful in the context of that class. Move to module level or autilsmodule only if it has clear reuse elsewhere.- Return or continue early to avoid deep nesting. Prefer comprehensions over loops that just build a list. Refactor code with more than three levels of nesting into helper functions.
- Use consistent naming. If a class is
FollowUpSuggestion, the variable should befollow_up_suggestion, notfollowup_suggestionorfollow_up_suggestions. - Sort
paramdeclarations alphabetically with a blank line between each. - Include
doc="""..."""on every public param, starting on a new line. - Ensure comments are about why and what must remain true, not what the syntax does. Good comments explain intent, constraints, workarounds, performance rationale, or API quirks. Avoid restating obvious code or narrating line-by-line. Keep them concise; over-explaining is also a smell.
- Compute derived values (ranges, extents, validation scans) once and reuse them; don't rescan the data in every method or on every render.
- Place internal
_-prefixed params after public params. Use a_-prefixed param (e.g._cache = param.Dict()) when the value needs to trigger watches or be serialized. Use a plain class/instance variable (e.g.self._cache = {}in__init__) for transient internal state that doesn't need param machinery.
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.
- 10d ago First seen · 100 lines · 38 tokens per session scan A 477cab7a4a27
cleanup is a skill published in the GitHub repository holoviz-dev/holoviz-skills (5 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 38 tokens to every session and 1,152 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…