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 rjroy/vibe-garden --skill ingestgit clone --depth 1 https://github.com/rjroy/vibe-gardenWrote 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/rjroy/vibe-garden/ingest)<a href="https://agentmods.dev/skills/rjroy/vibe-garden/ingest"><img src="https://agentmods.dev/badge/skills/rjroy/vibe-garden/ingest/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/rjroy/vibe-garden/ingest"><img src="https://agentmods.dev/badge/skills/rjroy/vibe-garden/ingest.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 | $0.00062 | $0.02085 |
| Opus 5.5 | $0.00025 | $0.00834 |
| Sonnet 5.5 | $0.00012 | $0.00417 |
| Haiku 4.5 | $0.00006 | $0.00209 |
Grade A, and why
ingest 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 17d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ingest
Distill useful project understanding into .lore/reference/. The reference is an aid to future work, not a contract or archive of project pages.
Keep only knowledge that code and tests do not adequately communicate and that materially helps a future change: rationale, real external constraints, relevant failed approaches, domain context, or useful current seams/coupling/refactoring guidance. Do not make a page just to summarize a feature, implementation, or source artifact. No useful candidate is a valid result; report that nothing was added.
Sources
Accept one or more paths from the user. Sources should live under .lore/; warn and skip anything else. For directories, walk supported .md and .html files. Historical brainstorms, designs, intents, research, and retros can suggest candidates. Plans, notes, and tasks (including under .lore/local/) are never reference sources: they may point to where to investigate, but do not lift their execution choices or claims into reference knowledge. If a directory contains no eligible sources, report that and continue.
Treat Markdown and HTML sources as equivalent knowledge inputs:
- Markdown files use YAML frontmatter when present, followed by Markdown body content.
- HTML files use
<meta name="...">fields when present,<title>as the title, and visible document body content as the source text. Ignore scripts, styles, and generated chrome that does not carry project knowledge.
Extraction
Read each eligible source in full. Treat it only as a lead, not proof of current behavior or authority over current user direction. For every candidate about current behavior, architecture, constraints, or extension seams, inspect relevant implementation and tests; correct, qualify, or discard claims they do not support. Historical rationale may be retained when useful even if code cannot prove it, but label it as historical rationale rather than current behavior or a mandate. The user's current direction prevails over any old artifact; call out a concrete safety or compatibility consequence when one matters, not documentary noncompliance.
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.
- 17d ago Changed · +4 lines · -24 tokens per session acf790daeca6
- 1mo ago First seen · 127 lines · 86 tokens per session scan A 86066667e04c
ingest is a skill published in the GitHub repository rjroy/vibe-garden (6 stars, last pushed 18d ago), licensed MIT. It adds 62 tokens to every session and 2,085 once invoked, about $0.0002 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.
Other skills, from other repositories
obsidian-cli
Obsidian vault integration via official CLI. Use when: capturing dev artifacts to Obsidian vault, searching vault for context, appending to daily note, managing tasks in vault. Not for: general note-taking without Obsidian (use regular files), browsing Obsidian docs (use agent-browser). Output: vault search results…
knowledge-structures
A system for organizing notes into navigation pages, synthesis notes, and small standalone notes. A map of content is a note that links together related notes around one topic.
pace-learn
A knowledge-management process that records reusable lessons and patterns from completed reviews, failed checks, rejected work, and signs of struggle.
pace-pulse
Auto-invoked during advance mode after extended work on same CR, at session start/end, or when rhythm anomalies are detected. NOT for metrics reports or retrospectives (use /pace-retro). NOT for project status overview (use /pace-status).
write-strategy
Interactive strategy document creation. Reads company context, researches competitors and market trends, then generates a comprehensive product strategy with vision, pillars, positioning, and metrics.
especialista-em-google-drive
Especialista em Google Drive. Use para organização, compartilhamento, permissões, Drives compartilhados, sincronização e colaboração no Google Drive. Palavras-chave: Google Drive, compartilhamento, permissões, Shared Drive, sincronização, organização, colaboração.