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 mblode/agent-skills --skill save-mdgit clone --depth 1 https://github.com/mblode/agent-skillsWrote 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/mblode/agent-skills/save-md)<a href="https://agentmods.dev/skills/mblode/agent-skills/save-md"><img src="https://agentmods.dev/badge/skills/mblode/agent-skills/save-md/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/mblode/agent-skills/save-md"><img src="https://agentmods.dev/badge/skills/mblode/agent-skills/save-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 48 This pattern attempts to override system instructions or ignore safety constraints. Without LLM analysis, manual review is recommended.Fix: Remove or rewrite any text that instructs the agent to ignore prompts, override safety rules, or trust unverified content. Ensure skill content cannot be injected to alter agent behavior.
- high YARA Match · line 48 YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
- high YARA Match · line 69 YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
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 | $0.00064 | $0.00537 |
| Opus 5.5 | $0.00026 | $0.00215 |
| Sonnet 5.5 | $0.00013 | $0.00107 |
| Haiku 4.5 | $0.00006 | $0.00054 |
Grade A, and why
save-md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Fetch with `curl -sSL -o <file> <url>` and read the file. Prefer a text endpoint over HTML chrome. A missing file with a named reason (paywall, login wall, bot check, private doc) beats a plausible one. What it actually says
Save as Markdown
Write the named source to a .md file, in full, and say where it is. The body comes from the fetched bytes, never from a summary.
- IS: one named source (URL, attachment, paste, YouTube video, Google Doc, PDF) to one
.mdin the user's tree. - IS NOT: answering from a fetch, summarizing into chat, crawling beyond the URLs named, or reconstructing a talk from memory.
Done
A .md exists where the user can open it, with this frontmatter, and reading it back returns the source's last paragraph:
---
title: "<title>"
source: "<URL or path>"
date: "<date -u +%Y-%m-%dT%H:%M:%SZ, run now>"
type: web | youtube | gdoc | sheet | slides | pdf | docx | tweet | conversation
---
Fetch with curl -sSL -o <file> <url> and read the file. Prefer a text endpoint over HTML chrome. A missing file with a named reason (paywall, login wall, bot check, private doc) beats a plausible one.
| File | Read when |
|---|---|
references/source-endpoints.md |
Always, for the URL's host: the tested endpoint, its curl line, the failure each one shows, and the opt-in fallback reader |
Gotchas
- WebFetch hands the page to a small model and returns its answer: a 1,200-word post came back as 120 words. Only
Content-Type: text/markdownunder 100K characters passes through intact. - oEmbed, GitHub
?raw=true, and Drive downloads redirect across hosts. Without-Lyou save an empty 301. - YouTube's "Sign in to confirm you're not a bot" is an IP block:
--cookies-from-browseron a laptop, stop in a sandbox. The description is not the talk. - An empty
.mdnext to a large PDF means a scan, not a blank document: render the pages and read them.
Maintenance only: evals/evals.json holds scenarios and routing prompts for anyone changing this skill.
What ships with it
2 files 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 Changed · -48 lines scan B → A d23c3ecd127d
- 1mo ago Changed · -11 lines · -114 tokens per session 900a86577e42
- 1mo ago First seen · 99 lines · 178 tokens per session scan B f4f0c385efc1
save-md is a skill published in the GitHub repository mblode/agent-skills (167 stars, last pushed today), licensed MIT. It adds 64 tokens to every session and 537 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
ingest
Bootstraps the memory store from existing material (repos, PDFs, transcripts). Use to seed memory from a project or document.
export
Bundle the canonical memory set into one structured Markdown document for external-LLM review, audit, or dashboard download.
dcf-model
Build discounted cash flow valuation workbooks in Excel.
comps-analysis
Build comparable-company valuation workbooks in Excel.
honcho
Configure and troubleshoot Honcho memory for Hermes.
lbo-model
Build leveraged buyout workbooks with IRR/MOIC in Excel.