wicked-garden-agentic-trust-and-safety SKILL.md skillA
wicked-garden-agentic-trust-and-safety from mikeparcewski/wicked-garden is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add mikeparcewski/wicked-garden/wicked-garden-agentic-trust-and-safety@git:20260914.823dcbcPins git:20260914.823dcbc in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- mikeparcewski/wicked-garden
- Path
skills/agentic/trust-and-safety/SKILL.md- Versions recorded
- 5
- Latest version
- git:20260914.823dcbc, committed Sep 14, 2026
- First version recorded
- Mar 22, 2026
- Size
- 6,190 bytes
- Repository stars
- 9
- License
- MIT
Trust, safety, and control patterns for production agentic systems with human-in-the-loop gates and guardrails. Use when: designing guardrails or human-in-the-loop gates for an agent, or hardening an agentic system against prompt injection.
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
# Trust and Safety
Essential patterns for building safe, trustworthy, production-ready agentic systems.
## Core Principles
1. **Least Privilege:** Agents have minimum necessary permissions
2. **Defense in Depth:** Multiple layers of safety checks
3. **Fail Safe:** Errors should fail toward safety, not capability
4. **Human Oversight:** High-stakes decisions require human approval
5. **Auditability:** All decisions and actions are logged and traceable
6. **Graceful Degradation:** System remains safe even when components fail
## Human-in-the-Loop Gates
### When to Add Human Gates
**Always require approval for:**
- Production data modifications (delete, update critical data)
- Financial transactions above threshold
- Communications to external parties
- Credential or security changes
- Irreversible operations
**Consider approval for:**
- First-time operations
- Operations outside normal patterns
- Low-confidence decisions
- Operations near resource limits
### Implementation Pattern
```python
async def execute_with_approval(action, threshold=0.8):
if action.confidence < threshold or action.is_high_stakes():
approval = await request_human_approval(action)
…Read the whole file at its exact version.
How to install
mdr add mikeparcewski/wicked-garden/wicked-garden-agentic-trust-and-safety@sha256:0798d8a5763e8af4The line at the top of this page pins the label git:20260914.823dcbc, 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_rphhgw4fg25upqs7)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Mar 22, 2026 and Sep 14, 2026, about one every 44 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260914.823dcbc latest | 2026-09-14 | 823dcbc | 6,190 B | A | view · diff |
| git:20260711.6a4c633 | 2026-07-11 | 6a4c633 | 6,208 B | A | view · diff |
| git:20260506.9dce6ee | 2026-05-06 | 9dce6ee | 6,186 B | A | view · diff |
| git:20260430.1fbddb4 | 2026-04-30 | 1fbddb4 | 6,121 B | A | view · diff |
| git:20260322.407f272 | 2026-03-22 | 407f272 | 6,092 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (6190 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 prompt-injection phrasing
- 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
mikeparcewski/wicked-garden · 9 stars · license MIT · pushed 2026-11-06 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_rphhgw4fg25upqs7 GET https://markdownregistry.com/api/v1/resolve?ref=mikeparcewski/wicked-garden/wicked-garden-agentic-trust-and-safety GET https://markdownregistry.com/api/v1/blob/0798d8a5763e8af4a009eea4ae4c518f7747ee894cf5bf61cfdeb01b9dcf294f
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.