Borrowing it
Nothing to install: this file belongs to edochi/mdvs. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/edochi/mdvs/main/.claude/skills/internalize/SKILL.mdgit clone --depth 1 https://github.com/edochi/mdvsWrote 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/edochi/mdvs/internalize)<a href="https://agentmods.dev/skills/edochi/mdvs/internalize"><img src="https://agentmods.dev/badge/skills/edochi/mdvs/internalize/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/edochi/mdvs/internalize"><img src="https://agentmods.dev/badge/skills/edochi/mdvs/internalize.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 | $0.00121 | $0.02874 |
| Opus 5.5 | $0.00048 | $0.01150 |
| Sonnet 5.5 | $0.00024 | $0.00575 |
| Haiku 4.5 | $0.00012 | $0.00287 |
Grade A, and why
internalize 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 10d 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 — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
internalize
Purpose
Agents write code faster than humans can absorb it, so understanding — not
writing — is the bottleneck. The durable reason to understand isn't to
verify (that's the validate step; agents increasingly self-verify) but to
participate: you can't steer the next iteration of a system you don't hold
in your head, and you can't feel ownership of decisions you never saw made.
This skill is a stateless transform: given a diff and the code it touches, it emits an interactive HTML explainer that rebuilds understanding fast. It does not decide how the diff was obtained, where the output lives, or when to run — those are the caller's concerns (see Inputs / Output).
Inputs (contract)
The caller supplies:
- A unified diff — from any source:
git diff(working tree / staged / a commit rangeA..B/ a PR), a raw patch piped in, or the output of a delta tool. The skill treats it as opaque unified-diff text; it does not care how it was produced. - Read access to the touched code — enough of the repo/tree to explore callers, callees, types, and config so the Background is real, not a paraphrase of the diff.
- (optional) Technical-depth hint —
line/module/architecture. Tunes how granularly the code is walked, not whether the wider objective is explained (that's always covered). If absent, infer from diff size (see Guidelines). - (optional) Output target — where to write the artifact. If absent, the skill writes to a scratch path and returns it; deciding the durable home is the caller's job.
If invoked ad-hoc with none of these, derive a convenience default (e.g.
git diff of the current branch vs its merge-base) — but treat that as a
fallback, not part of the skill's responsibility.
Guidelines
Grounding — this skill builds ownership, so it must never manufacture it. Every factual claim about the existing system (Background) and every asserted rationale (Decision log) must trace to code you actually read, or to the diff, commits, or comments themselves. Tag each non-obvious claim by its epistemic tier: (stated) — present in the diff / comment / commit message; (inferred) — your read of the evidence, cite the file or symbol that supports it; (open question) — genuinely unknowable from what you can see, flag it rather than guess. Never present an inference or a guess as a stated fact. An explainer that fabricates the "why" is worse than none — the reader ends up owning a fiction and steers the next iteration wrong.
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.
- 10d ago First seen · 210 lines · 121 tokens per session scan A 92b423a411d3
internalize is a skill published in the GitHub repository edochi/mdvs (27 stars, last pushed 11d ago), licensed MIT. It adds 121 tokens to every session and 2,874 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 2026-10-05.
Other skills, from other repositories
processing-markdown
Processes Markdown files using mq, a jq-like query language for Markdown. Use when the user mentions Markdown processing, content extraction, document transformation, or mq queries.
larksnap-fetch
A bridge for downloading Feishu/Lark documents or ordinary webpages into local files, with options such as Markdown, HTML, or PDF. Feishu, also called Lark, is a workplace collaboration platform.
hyalo-okf
Maintain Open Knowledge Format vaults with Hyalo's OKF index, log, and conformance commands.
wechat-article-to-markdown
A skill that fetches articles from WeChat Official Accounts and converts them to Markdown, a plain-text format commonly used for documentation.
haypile
Search the user's local documents (PDF, docx, pptx, markdown, text, HTML, mbox email) through Haypile, a local search daemon. Use when the user asks what their files say, wants passages from contracts, papers, or notes, or wants a folder indexed for search. Every result carries a file and page citation. Runs entirely…
pdftomarkdown
Converts PDF documents, Word files (.docx), CSV spreadsheets, and YouTube video URLs into clean, token-efficient Markdown with smart table and heading structure detection.