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 technicalpickles/pickled-claude-plugins --skill vault-capturegit clone --depth 1 https://github.com/technicalpickles/pickled-claude-pluginsWrote 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/technicalpickles/pickled-claude-plugins/vault-capture)<a href="https://agentmods.dev/skills/technicalpickles/pickled-claude-plugins/vault-capture"><img src="https://agentmods.dev/badge/skills/technicalpickles/pickled-claude-plugins/vault-capture/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/technicalpickles/pickled-claude-plugins/vault-capture"><img src="https://agentmods.dev/badge/skills/technicalpickles/pickled-claude-plugins/vault-capture.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 | $0.00113 | $0.01680 |
| Opus 5.5 | $0.00045 | $0.00672 |
| Sonnet 5.5 | $0.00023 | $0.00336 |
| Haiku 4.5 | $0.00011 | $0.00168 |
Grade A, and why
vault-capture 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 1mo 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 — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vault Capture
Turn a source (or a raw idea) into one Obsidian note file, already shaped for Josh's
pickled-knowledge vault - correct filename, correct frontmatter, correct body shape, and a
best-guess at where it'll eventually live. He should be able to save the file straight into his
vault with no renaming, no reformatting, no hunting down the source URL after the fact.
This is a chat-only rewrite of a more capable version that runs inside Claude Code (searches the vault first, checks for duplicates, auto-links related notes, appends a daily-note breadcrumb, actually files the note). None of that is reachable from here - no filesystem, no vault search index. Say so plainly once, up front, rather than quietly skipping it:
"Heads up: I can't search your vault from here, so I'm not checking for an existing note or suggesting related links - worth a quick check yourself before you file this."
Step 1: Get the real source - never invent one
- A URL was given: fetch and read it for real (browse it). Never write the note from the URL text alone, a guess at what the page probably says, or general knowledge about the topic. If the page can't be reached, say so and stop - don't write a plausible-sounding note anyway.
- A tweet/thread: read the actual tweet content if it can be fetched. If not reachable, say so rather than reconstructing it from context.
- No URL, just an idea or topic: there is no source to fetch. Don't backfill one from
training knowledge and present it as researched. Write the note from what Josh actually said,
and leave
sourceoff the frontmatter entirely. - Several sources at once ("make notes for each of these"): handle them one at a time, in order, and produce one file per source. There's no subagent fan-out available here.
Step 2: Write the frontmatter
---
status: seedling
tags: [lowercase, hyphenated, no-leading-hash]
created: 2026-09-11
source: https://example.com/the-actual-page
---
status: seedlingalways, for a fresh capture. This is the note's maturity, not a workflow state - don't writefiled,active,incubating, etc. Those belong to notes that have already been routed inside the vault, which this one hasn't.tags: a short list, lowercase, hyphenated, no#prefix (a bare#at the start of a YAML flow-sequence item is a comment and silently corrupts the frontmatter). Pick 2-4 topic tags from what the note is actually about. references/jd-map.md is useful here too - a tag that matches how the vault already splits a topic (e.g.rubyvs a genericprogramming) ages better.created: today's date,YYYY-MM-DD.source: the real URL/origin actually fetched in Step 1. Omit the field entirely for a sourceless idea capture - don't writesource: noneor similar.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 125 lines · 113 tokens per session scan A 38c3236c08f0
vault-capture is a skill published in the GitHub repository technicalpickles/pickled-claude-plugins (11 stars, last pushed yesterday), licensed MIT. It adds 113 tokens to every session and 1,680 once invoked, about $0.0005 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 2026-09-17.
Other skills, from other repositories
format-converter-brief
Produces a clear, step-by-step brief explaining exactly how to convert a finished article or document from one format (Word, plain text, InDesign export) into a specified target format (Markdown, HTML, or structured plain text), preserving all editorial elements correctly.
doc-writing
A guide for creating and maintaining technical documentation such as READMEs, API references, user guides, and office documents.
pptx-creating
A skill for creating, editing, and analyzing PowerPoint presentations in PPTX format. It covers slides, templates, comments, speaker notes, and presentation structure.
pydicom
Reads, inspects, writes, transforms, and preflights local DICOM datasets and pixel data. Applies to DICOM metadata, transfer syntaxes, compression plugins, frames, private elements, JSON, and bounded de-identification review.
treatment-plans
Formats and structurally validates local treatment-plan documentation after clinical decisions have already been supplied and verified by authorized licensed professionals. Use for source traceability, clinician-authored intervention records, goals and checkpoints, shared-decision records, reconciliation handoffs, and…
venue-templates
Prepares journal manuscripts, conference papers, research posters, and grant documents using venue-specific formatting guidance and bundled LaTeX scaffolds. Supports selecting an official template, checking current page or anonymity rules, adapting academic writing to a venue, or inspecting a submission PDF.