untrusted-input-boundary-review SKILL.md skillB
untrusted-input-boundary-review from gsa-tts/agentic-coding-patterns is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add gsa-tts/agentic-coding-patterns/untrusted-input-boundary-review@v1.0.0Pins v1.0.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- B, 16 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- gsa-tts/agentic-coding-patterns
- Path
.agents/skills/untrusted-input-boundary-review/SKILL.md- Versions recorded
- 3
- Latest version
- v1.0.0, committed Aug 20, 2026
- First version recorded
- Jul 2, 2026
- Size
- 11,512 bytes
- Repository stars
- 16
- License
- CC0-1.0
Map an agent's trust boundaries and review how it handles UNTRUSTED input (issue/PR text, tool/MCP output, web content, agent-to-agent messages) for prompt-injection and tool-poisoning — describing attack CLASSES, never shipping a working exploit.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Skill: Untrusted Input Boundary Review Map the **trust boundaries** of an agent, workflow, or automation, then review how it handles **untrusted input** — issue and PR text, tool and MCP output, web content, and agent-to-agent messages. The goal is to confirm untrusted content is treated as **data**, never as instructions the agent obeys. > **Self-limiting:** This skill reviews *trust-boundary handling*. It describes > injection and tool-poisoning **classes conceptually** and **never** crafts, > tests, or ships a working attack string or jailbreak prompt. It does not set > policy — authoritative prompt-injection defense lives in the > [agentic-coding-playbook](https://github.com/GSA-TTS/agentic-coding-playbook), > and this skill defers all rules to it. It is not a substitute for > app-level code review or a CI-permissions audit. ## When to Use - Reviewing an agent, prompt, or workflow that consumes external content - The agent reads issue/PR bodies, comments, web pages, or file contents - The agent calls tools or MCP servers and acts on their output - Multiple agents exchange messages and act on each other's text …
Read the whole file at its exact version.
How to install
mdr add gsa-tts/agentic-coding-patterns/untrusted-input-boundary-review@sha256:e781cc524639dd4fThe line at the top of this page pins the label v1.0.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_qhvcwxak7qhk2t5n)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Jul 2, 2026 and Aug 20, 2026, about one every 24 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.0.0 latest | 2026-08-20 | ae5b0dd | 11,512 B | B | view · diff |
| v1.0.0 | 2026-07-28 | ea6ea67 | 11,509 B | B | view · diff |
| v1.0.0 | 2026-07-02 | c736287 | 11,080 B | B | view |
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 (11512 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
Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.
Source
gsa-tts/agentic-coding-patterns · 16 stars · license CC0-1.0 · pushed 2026-10-19 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_qhvcwxak7qhk2t5n GET https://markdownregistry.com/api/v1/resolve?ref=gsa-tts/agentic-coding-patterns/untrusted-input-boundary-review GET https://markdownregistry.com/api/v1/blob/e781cc524639dd4ff2e53946237710aaa28faf6a0ae38a17765652b1d8c1f3d4
Your AI assistant 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.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.