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/jpd44/mesh44-plugins/capturenpx skills add jpd44/mesh44-plugins --skill capturegit clone --depth 1 https://github.com/jpd44/mesh44-pluginsWhat 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 | $0.00058 | $0.01087 |
| Opus 5 | $0.00029 | $0.00544 |
| Sonnet 5 | $0.00012 | $0.00217 |
| Haiku 4.5 | $0.00006 | $0.00109 |
Grade A, and why
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 yesterday.
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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/devkb:capture — save session ideas to your Obsidian knowledge base
Turn ideas that came up while working into durable, linkable notes in the user's Obsidian vault. Two modes:
- A specific idea — the user points at one thing ("save that idea about caching the cost snapshot"). Capture just that.
- Sweep the session — the user asks to capture everything worth keeping ("grab any ideas from this session"). Scan the conversation and pull out the genuinely new, reusable ideas.
Curate. An idea worth a note is a reusable insight, design direction, or thing-to-build — not a task you just finished, a bug you fixed, or routine status. When in doubt, prefer fewer, sharper notes over many thin ones.
Where the vault is
Read the vault path from ~/.config/mesh44/config.json:
{
"obsidian": {
"vault": "/Users/<you>/Documents/Obsidian Vault",
"folder": "devkb"
}
}
- Notes go in a
devkbfolder inside the vault. Ifobsidian.folderis unset, default todevkb; if theobsidiankey is missing entirely, default to~/Documents/Obsidian Vaultanddevkb— and tell the user you're using the default and they can setobsidian.vault/obsidian.folderto change it. - If the vault directory doesn't exist, ask the user for the vault path rather than guessing. Never write outside the vault.
- Create
<vault>/<folder>/if it doesn't exist yet.
Where in the folder — general vs project
The <folder>/ is organised into two kinds of subfolder:
<folder>/general_learnings/— cross-project, reusable insights: a pattern, a gotcha, a principle, a technique that would apply on any project.<folder>/<project>/— anything specific to the project this session is about (an idea for this app, a decision, a thing to build here).
For each idea, decide which it is. Derive <project> as a clean slug from the repo/cwd — e.g. a session in www.mesh44.com is project mesh44 (drop a leading www. and a trailing TLD-like suffix). If a matching project folder already exists, reuse it exactly rather than making a near-duplicate. When genuinely unsure, prefer the project folder — it's easy to promote to general_learnings/ later.
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.
- yesterday First seen · 80 lines · 58 tokens per session scan A 6ee199152abe
capture is a skill published in the GitHub repository jpd44/mesh44-plugins (2 stars, last pushed 1mo ago), licensed MIT. It adds 58 tokens to every session and 1,087 once invoked, about $0.0003 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.