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 0ics-srls/lsai-xmp4.public --skill xmp4git clone --depth 1 https://github.com/0ics-srls/lsai-xmp4.publicWrote 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/0ics-srls/lsai-xmp4.public/xmp4)<a href="https://agentmods.dev/skills/0ics-srls/lsai-xmp4.public/xmp4"><img src="https://agentmods.dev/badge/skills/0ics-srls/lsai-xmp4.public/xmp4/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/0ics-srls/lsai-xmp4.public/xmp4"><img src="https://agentmods.dev/badge/skills/0ics-srls/lsai-xmp4.public/xmp4.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 | $0.00328 | $0.05801 |
| Opus 5 | $0.00164 | $0.02900 |
| Sonnet 5 | $0.00066 | $0.01160 |
| Haiku 4.5 | $0.00033 | $0.00580 |
Grade A, and why
xmp4 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 11d 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 — 441 lines — stays where its author put it; the contents beside it link to each section on GitHub.
xmp4 — Code Intelligence for Third-Party Libraries
xmp4 is an MCP server that answers semantic code questions over pre-indexed OSS libraries. Reach for it first when the user wants to understand how an unfamiliar library is actually used in practice — it is far cheaper and more focused than grepping source archives or reading READMEs.
The primary value is this: instead of guessing from signatures or docs, you can instantly find the real tests and call sites that exercise a symbol, then read them. That is where idiomatic usage patterns actually live.
Project id contract — read this first
Every xmp4 data tool takes a project argument. That argument is a
project id produced by xmp4's indexing pipeline — not a GitHub URL,
not an npm slug, not a package name.
The shape is derived from the upstream git coordinates, typically
<org>/<repo> for a single-project repo (e.g. django/Django,
flask/Flask, fastapi/fastapi), or <org>/<subproject> for a
monorepo member (e.g. graphql-platform/HotChocolate.Core,
runtime/System.Collections.Tests). A reasoned guess often works on
the first try.
But case is preserved as the indexer stored it (django/Django,
not django/django), monorepos may use curated naming
(aws-sdk-go-v2-curated/aws-sdk-go-v2, not aws-sdk-go-v2/*), and
the segmentation may not be obvious. If your guess is rejected
with project_id_not_found, do not iterate variations. Instead:
xmp4_projects(query="<keyword>") # one call — returns authoritative ids
Copy an id verbatim from the result, then retry your original call with it.
Session reuse — do not re-discover
Once xmp4_projects returns an id, keep it in your working memory
for the rest of this conversation. For every subsequent xmp4 tool
about the same library, reuse the id verbatim. Re-discover only when
you switch to a different library.
Calling xmp4_projects for the same keyword twice in the same session
wastes round-trips on the server we operate for you. If your client
supports per-workspace persistent memory, you can also save discovered
ids there to skip discovery in future sessions.
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.
- 11d ago First seen · 441 lines · 328 tokens per session scan A f7d74a83e96a
xmp4 is a skill published in the GitHub repository 0ics-srls/lsai-xmp4.public (0 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 328 tokens to every session and 5,801 once invoked, about $0.0016 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
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…