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 ManceRayder42/jarvis-os --skill knowledge-packgit clone --depth 1 https://github.com/ManceRayder42/jarvis-osWrote 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/mancerayder42/jarvis-os/knowledge-pack)<a href="https://agentmods.dev/skills/mancerayder42/jarvis-os/knowledge-pack"><img src="https://agentmods.dev/badge/skills/mancerayder42/jarvis-os/knowledge-pack/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/mancerayder42/jarvis-os/knowledge-pack"><img src="https://agentmods.dev/badge/skills/mancerayder42/jarvis-os/knowledge-pack.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 | $0.00072 | $0.00760 |
| Opus 5.5 | $0.00029 | $0.00304 |
| Sonnet 5.5 | $0.00014 | $0.00152 |
| Haiku 4.5 | $0.00007 | $0.00076 |
Grade A, and why
knowledge-pack 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 15d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge pack
A knowledge pack is a folder of markdown holding what is worth knowing about one subject, arranged so a later session loads only the part it needs. A single rule file has to be either short enough to always load or deep enough to be worth writing. A pack is both: a tiny index that is always read, and large files read only when the task touches them.
When it earns its place
Only when something keeps being re-derived: a design system's rules, a compliance regime, a vendor's API quirks, a market you serve. If nothing re-derives it, a single hub note is enough. Check the hub first for an existing pack or note.
Layout (in the hub, e.g. <hub>/knowledge/<subject>/)
index.md the spine and routing table; always read first
01-<topic>.md one file per sub-domain, numbered in reading order
NN-<topic>.md
sources.md every source, with its tier and how to re-fetch it
File contract
Frontmatter on every file:
---
name: <kebab-case, matches filename>
description: <one line: what this file knows>
read-when: <a concrete trigger, e.g. "before choosing a payment flow">
tier-key: <what A/B/C mean in this file>
---
Body rules:
- Operational register: rules that can be applied and checked. No essays.
- Numbers over adjectives. If a rule can carry a number, it must.
- Every claim carries a source URL and an evidence tier: A = primary source or standards body, B = replicated practitioner consensus, C = single unsourced claim.
- Never invent a number. A widely repeated statistic you cannot source is recorded as folklore; that is a finding.
- Say where evidence is thin. Flag contradictions between good sources; do not average them.
Procedure
- Scope the subject and split it into 5 to 9 sub-domains. Fewer files are too coarse to load selectively; more and the router is not memorable.
- Fan out one research sub-agent per sub-domain, in a single message. Give each: its sub-domain, the file contract above, the target filename, and a deliverable shape (the file, the evidence for its key claims, and a "What I did not check" section). They need no coordination; the contract is the coordination.
- Review each file against the contract: sourced, tiered, numeric, honest about gaps. Send back any that fail, with the failed rule named.
- Write
index.mdlast: a short summary of the subject, the routing table (question to file), and the list of files with theirread-whenlines. - Add one line to the hub's
MEMORY.mdpointing at the pack'sindex.md.
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.
- 15d ago First seen · 72 lines · 72 tokens per session scan A 4faf4d7df379
knowledge-pack is a skill published in the GitHub repository ManceRayder42/jarvis-os (5 stars, last pushed 15d ago), licensed MIT. It adds 72 tokens to every session and 760 once invoked, about $0.0003 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 2026-10-01.
Other skills, from other repositories
memory-to-skill
Turn workflows from your MemSearch memory into reusable skills. Use when the user asks to make/create/extract/distill a skill from what they just did or from past work, review skill candidates, install a distilled skill, or 'turn this into a skill'. Manages MemSearch procedural-memory candidates under…
learn
Must be used near the end of any non-trivial turn that produced potentially reusable tools, guidance, errors, workarounds, or workflows, so those lessons are saved for future turns.
adapt-memory
Mirror a just-saved native memory into the shared evolve store so it becomes shareable and auditable.
recall
Must be used at the start of any non-trivial task involving code changes, debugging, repo exploration, file inspection, or environment/tooling investigation to surface stored guidance before analysis or tool use.
setup-brain
Set up or upgrade an AI-powered Obsidian vault. Interviews you, builds your vault structure (or works with what you already have), creates your CLAUDE.md memory file, installs tools, and gets you journaling — all in one conversation. Also has a repair/upgrade path for existing users.
token-check
Estimate how many tokens a pasted text, an attached file or the conversation so far takes, what it means for the context window and usage limits, and how to cut it. Use when the user asks how big something is in tokens, why they keep hitting limits, or whether a file is too large to send.