Borrowing it
Nothing to install: this file belongs to acaprino/daodan. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/acaprino/daodan/master/.agents/skills/custom-plugin-refresh/SKILL.mdgit clone --depth 1 https://github.com/acaprino/daodanWrote 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/acaprino/daodan/custom-plugin-refresh)<a href="https://agentmods.dev/skills/acaprino/daodan/custom-plugin-refresh"><img src="https://agentmods.dev/badge/skills/acaprino/daodan/custom-plugin-refresh/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/acaprino/daodan/custom-plugin-refresh"><img src="https://agentmods.dev/badge/skills/acaprino/daodan/custom-plugin-refresh.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 | $0.00177 | $0.03567 |
| Opus 5.5 | $0.00071 | $0.01427 |
| Sonnet 5.5 | $0.00035 | $0.00713 |
| Haiku 4.5 | $0.00018 | $0.00357 |
Grade A, and why
custom-plugin-refresh 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 — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Custom plugin maintenance
A "custom plugin" has hand-authored or research-grounded content without a copied upstream source to re-pull. Copied-source targets in the upstream-sync table use that skill for upstream comparison. Inspiration-only rows retain provenance without excluding the original local plugin from this refresh method: both codebase-map rows are inspiration-only, so refresh its local content here. An inspiration row that records an upstream commit is a watch: at each refresh, read the upstream diff of the named path since that commit (gh api repos/<owner>/<repo>/compare/<sha>...HEAD), decide which ideas to adopt, rewrite them in local words, then advance the recorded commit. Never paste upstream text. Preserve existing copied-source tracking and history; neither route authorizes new vendoring.
Custom plugins decay differently than vendored ones. There is no upstream commit to diff against. Versions, framework recommendations, breaking-change notes, and "current as of 2026" claims become stale silently. The maintenance protocol below is the antidote.
Refresh the hand-authored kernel under plugins/<name>/: plugin.toml, roles/,
workflows/ with their TOML sidecars, skills/, contracts/ and policies/.
Host mechanisms live under adapters/. Exports and root marketplace catalogs are
generated; correct their source and rebuild rather than editing package copies.
Freshness risk classes
Classify each plugin into one of four classes. The class determines refresh cadence and triage priority.
| Class | What it tracks | Typical cadence | Examples |
|---|---|---|---|
| Very fast | Versions bump every few months; breaking changes are common; ecosystem reshuffles | Every 3 months | rag-development (embedding models, rerankers, vector DBs), digital-marketing/ga4-implementation (Consent Mode, GA4 events), react-development (React 19, Vercel guidance) |
| Fast | Framework releases 2-3x per year; APIs evolve | Every 6 months | libgdx-development, opentelemetry, tauri-development, stripe (API additions, webhook event types), grabber-development (anti-bot vendor moves), browser-extensions, pwa-expert (browser version churn, WebKit feature rollout, framework PWA library churn), text-humanizer (Wikipedia's signs page and the upstream humanizer both change monthly), business (its privacy-doc-generator quotes app store rules and EDPB guidance; see its refresh note below) |
| Moderate | Major releases ~yearly; breaking changes rare | Every 12 months | trading-broker-integration, csp (OR-Tools), python-development, typescript-development, messaging (RabbitMQ majors), obsidian-development, abstraction-architect (theory is stable; URL list in further-reading.md decays on a yearly cadence) |
| Slow | Workflow knowledge that ages by behavior change, not version bumps | Opportunistic; review only when symptoms appear | project-delivery, project-lifecycle, work-controller, senior-review, project-knowledge, team pipeline workflows, marketplace-ops, system-utils, learning, research, clean-code, codebase-map, platform-engineering, testing methodology, xterm, app-analyzer |
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 Changed 41886d816938
- 11d ago Changed · +18 lines 769a9d180ff9
- 12d ago Changed c48697361f3f
- 13d ago Changed · +1 lines e19d3a60949d
- 14d ago Changed · +4 lines 98d605f9e091
- 23d ago Changed 8c8928c87124
- 24d ago Changed · +5 lines · +5 tokens per session a75c3da45c87
- 1mo ago Changed · +6 lines b86877867231
- 1mo ago Changed 0adc6127142f
- 2mo ago First seen · 116 lines · 172 tokens per session scan A 5ac2890530db
custom-plugin-refresh is a skill published in the GitHub repository acaprino/daodan (9 stars, last pushed 4d ago), licensed MIT. It adds 177 tokens to every session and 3,567 once invoked, about $0.0007 per session on Opus 5.5. 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.
custom-plugin-refresh compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| custom-plugin-refreshacaprino/daodan | Skill | 9 | 4d ago | MIT | A | 177 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |