agent-guardrails skillB
agent-guardrails is agent-read markdown (skill) from akillness/jeo-skills: Layered agent safety — prompt-injection defense, data exfiltration prevention, tool-misuse blocking via NeMo Guardrails + Lakera / LLM Guard / Llama Guard. Use before any agent reads untrusted input or executes tools with real-world side effects..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Agent Guardrails ## Overview A single LLM call cannot be trusted to police itself. Production agents need **layered** defenses: input filters, output filters, tool allow-lists, and behavioral policies. This skill picks the libraries and wires them in at the right points. ## When to use - Agent reads untrusted input (web pages, emails, user uploads) - Agent has tools that touch real systems (DB, API, money, send messages) - Compliance requires PII redaction or content policy enforcement - You've seen prompt injection or data exfiltration attempts in logs - LLM-as-judge alone has been bypassed in red-team tests ## Threat model (2026) | Threat | Example | Defense layer | |--------|---------|---------------| | Direct prompt injection | "Ignore previous instructions" | Input filter | | Indirect injection | Malicious instruction in a fetched web page | Provenance + scoped tools | | Data exfiltration | Agent emails secrets to attacker | Output filter + tool allow-list | | Tool misuse | Agent calls `delete_db` on bad reasoning | Policy gate + HITL | | Jailbreak | Roleplay bypass | Llama Guard / NeMo Colang | …
Read the whole file at its exact version.
How to install
mdr add akillness/jeo-skills/agent-guardrails@git:20260919.9f2a908mdr add akillness/jeo-skills/agent-guardrails@sha256:d156a519f2c1e255Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_5d34ta2kxu5mo745)
1 badge views in 30 days
Versions
Audit of the latest version
- fail: No prompt-injection phrasing (matched: Ignore previous instructions)
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (4064 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
akillness/jeo-skills · 47 stars · license none · pushed 2026-09-22 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_5d34ta2kxu5mo745 GET https://markdownregistry.com/api/v1/resolve?ref=akillness/jeo-skills/agent-guardrails GET https://markdownregistry.com/api/v1/blob/d156a519f2c1e2552adf04e766974d5ee7176c43e83594fbabbdfa1183f94e85
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.