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/cloud-forest-analytics/evergreen/weekly-seedlingnpx skills add Cloud-Forest-Analytics/evergreen --skill weekly-seedlinggit clone --depth 1 https://github.com/Cloud-Forest-Analytics/evergreenWrote 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/cloud-forest-analytics/evergreen/weekly-seedling)<a href="https://agentmods.dev/skills/cloud-forest-analytics/evergreen/weekly-seedling"><img src="https://agentmods.dev/badge/skills/cloud-forest-analytics/evergreen/weekly-seedling.svg" alt="Measured on agentmods" 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 | $0.00142 | $0.01750 |
| Opus 5 | $0.00071 | $0.00875 |
| Sonnet 5 | $0.00028 | $0.00350 |
| Haiku 4.5 | $0.00014 | $0.00175 |
Grade A, and why
weekly-seedling 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The weekly-seedling
A once-a-week, self-contained HTML newsletter surveying the state of the art in this repo's own domain — relevant releases, research, techniques, and an opinionated read on what's worth adopting. It is the renewal artifact of the maintenance loop: research -> an editorial burn (default-cut review) -> this newsletter -> a few board items -> measure.
The through-line — read first: every issue answers ONE question: "what should this repo adopt, borrow, watch, or skip this week?" It is the repo's scouting dispatch for its own roadmap, not a trade magazine. If an item doesn't change what this repo could build or borrow, it doesn't run.
Cadence: weekly. One issue per ISO week; the issue id IS the ISO YYYY-WW (e.g. 2026-25).
An intra-week revision appends .1, .2 and keeps the same identity. Output: a fully self-contained
.html (inline CSS, no CDN/JS/web-font fetches) saved under newsletter/issues/ and filed as one
GitHub issue.
0 · Infer the domain (do this first, every run)
This skill is repo-agnostic. Before researching, establish what this repo is about from its own
artifacts — README.md, CLAUDE.md/AGENTS.md, the top-level package manifest, and the dominant
languages/dirs. Write a one-sentence domain statement ("this repo is a that does "). Every
research query and every verdict is scoped to THAT domain. Never drift into general AI/tech news.
1 · Research (fresh every week — GitHub-first, not press-release-first)
- Source from REPOS, not corporate blogs. Lead with GitHub repo search,
awesome-*lists, papers-with-code, and actual implementations you can read in the repo's domain. Cite the repo URL- stars/recency. Minimize vendor/lab press announcements; when one matters, reframe it as "what's the open-source equivalent, and what could we build or borrow?"
- Domain-specific. Only developments that change a technique, library, or approach relevant to THIS repo. A general model benchmark is out of scope unless it changes something here.
- Diff against last issue (read the previous
newsletter/issues/file). Cite every claim; adversarially sanity-check — no hallucinated repos, stars, or results. - Watchlist feed (read the board, not just the web). Pull open waves the team is already tracking:
Eachgh issue list --label watchlist --state open --json number,title,body gh issue view <N> --json comments --jq '.comments[-1].body' # latest scan digest per wavewatchlist-labeled issue is one wave; its newest comment is that wave's latest digest. These feed the Watchlist section below — not a fresh web hunt.
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 · 112 lines · 142 tokens per session scan A a2a6e4c1ae4d
weekly-seedling is a skill published in the GitHub repository Cloud-Forest-Analytics/evergreen (1 stars, last pushed 2mo ago), licensed MIT. It adds 142 tokens to every session and 1,750 once invoked, about $0.0007 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
agents-maintenance
Use when working in monorepos with multiple apps/packages, when CLAUDE.md or AGENTS.md files need validation, when detecting broken links, redundant or conflicting instructions, vague instructions, incomplete coverage, or inconsistent formatting - validates structure, links, and instruction quality with interactive…
i18n-maintenance
Use when working with internationalization files, translation keys, or multi-language support - validates translation coverage, detects missing keys, finds unused translations, validates code↔JSON sync, and ensures consistency across all locales. Triggers on i18n, translation, locale, intl, multi-language, missing…
pr-lifecycle
Full PR lifecycle management: list, analyze, validate, and merge pull requests. Auto-detects validation gates per package and integrates a code-review pass for code analysis. Use when: managing PRs, checking PR status, reviewing open PRs, merging PRs, triaging PRs, cleaning up stale PRs, validating PR readiness, or…
prompt-maintenance
Maintain prompt templates with dependency tracking, documentation, and validation. Use when auditing prompts, generating catalog, or checking template structure.
readme-maintenance
This skill should be used when the user asks to "audit READMEs", "update documentation", "check README consistency", "find missing READMEs", "generate README", "validate docs", or mentions README/documentation maintenance for packages and apps.
Buttondown Newsletter
This skill should be used when the user asks to "manage newsletters", "create an email draft", "list my drafts", "schedule a newsletter", "get email analytics", "check newsletter stats", "send a newsletter", "create buttondown draft", or mentions Buttondown newsletter management. Provides tools for creating…