AgentGuards
AgentGuards v1.0.4
Security scans for AI agents
From the developer
Declared capabilities
- Scan text for prompt injection and jailbreaks
- Detect leaked secrets and personal data
- Check shell commands before running them
Declared limitations
- Asks for an action (delete files). AgentGuards only checks commands and text and never runs or deletes anything, so it should not be invoked to do this.
- A general security question with nothing to scan; the assistant answers from its own knowledge.
- How-to advice about credentials, not a scan of any content.
Not specified in the available declarations: pricing & account requirements.
Onboarding example
Is this email safe to act on, or does it contain a prompt injection?
Sources & dates
Developer declarations; not independently tested. Listing last checked Nov 6, 2026. Package observed Nov 6, 2026.
Description
From the marketplace listing
AgentGuards checks the things an AI agent is about to trust or run. - Scan untrusted text (web pages, emails, documents, issues, other tools' output) for prompt injection, jailbreak attempts, leaked secrets and personal data before acting on it. - Check a shell command before running it: destructive deletes, credential reads, data exfiltration, force pushes and infrastructure changes are flagged with a risk level and a reason. - Scan source code for security issues and hardcoded secrets before committing or running it. Who it's for: developers using ChatGPT or Codex with untrusted inputs or shell access. Requires an AgentGuards account (you sign in when you connect). Scan results also appear in your AgentGuards dashboard. Limitations: the tools scan only what they are given, when they are called. They do not intercept anything automatically, and they never run, change or delete anything themselves.
Language: English · Automatically detected from descriptions.
Matches for “developers”
Exact text from the indicated source. A mention alone does not establish support for your task.
Publisher full description
AgentGuards checks the things an AI agent is about to trust or run. - Scan untrusted text (web pages, emails, documents, issues, other tools' output) for prompt injection, jailbreak attempts, leaked secrets and personal data before acting on it. - Check a shell command before running it: destructive deletes, credential reads, data exfiltration, force pushes and infrastructure changes are flagged with a risk level and a reason. - Scan source code for security issues and hardcoded secrets before committing or running it. Who it's for: developers using ChatGPT or Codex with untrusted inputs or shell access. Requires an AgentGuards account (you sign in when you connect). Scan results also appear in your AgentGuards dashboard. Limitations: the tools scan only what they are given, when they are called. They do not intercept anything automatically, and they never run, change or delete anything themselves.
Changes
No changes recorded yet. Follow this plugin to keep track of future changes.
First seen Nov 5, 2026. Last seen Nov 6, 2026.
Similar plugins
Find competitors to follow as you develop your plugin.
-
Test your AI agents for security and safety weaknesses, and re-verify the fixes.
Shared terms: security, agent, prompt, code
-
Scan agent skills and plugins locally for reviewable risk patterns before installation.
Shared terms: scan, injection, prompt, risk
-
Audit AI coding sessions and assistant configs with the locally installed skarn CLI: leaked credentials and risky hooks or MCP servers, r...
Shared terms: security, attempt, command, credential
-
VIDOC Security Review by Vidoc Security Lab: review source code for security issues, validate suspected findings, and prepare reports wit...
Shared terms: security, developer, issue, vulnerabilitie
-
Rollcage checks shell commands, file paths and URLs against the rules of the open-source Rollcage Guard before an AI coding agent runs th...
Shared terms: agent, command, destructive, force
Keywords
Files & skills
File archives
Release notes
Three read-only security tools: scan untrusted text (web pages, emails, documents, tool output) for prompt injection, jailbreak attempts, leaked secrets and personal data; check shell commands for risk before running them; and scan code for vulnerabilities and hardcoded secrets. Connect with an AgentGuards account.
Declared in the saved package. Remote tools may change independently.
Package details
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
- Package author
- AgentGuards
- Keywords
- See keywords
- Commerce declaration
- Does not support commerceThis does not establish whether access is free or paid.
- Review scenarios
- 5 positive · 3 negativeScenario counts, not test results.
Declared capabilities
- Scan text for prompt injection and jailbreaks
- Detect leaked secrets and personal data
- Check shell commands before running them
- Scan code for vulnerabilities and secrets
Package observed Nov 6, 2026.
Technical details
- First seen
- Nov 5, 2026 · 12:00 UTC
- Last seen
- Nov 6, 2026 · 18:00 UTC
- Collection status
- Collected
plugin_asdk_app_6ad5d8f54ea881918aea2081f5ccc067
Download plugin data (JSON)Plugin details: freshly prepared. Prepared . Refreshed when plugin data changes or after 6 hours. Your account and Follow status are live.