Policy pack · Security
OWASP agentic default policy pack
Default security policy set for new Kyvvu installations, providing baseline coverage of the OWASP Top 10 for Agentic Applications (2026) at agent development time.
What's in it
8 policies, 5 of them blocking.
Critical policies block the step before it runs. High, medium and low ones let it run and open an incident.
| Policy | Checked | Severity | Rule |
|---|---|---|---|
| Agent must declare a substantive purpose | At registration | high | field_matches_regex |
| Agent must declare a tool allowlist | At registration | high | field_not_empty |
| Agent must declare a valid risk classification | At registration | critical | field_in_list |
| Tool calls must be in the declared allowlist | Before every step | critical | step_name_in_allowlist |
| Code execution requires a preceding gate | Before every step | critical | step_requires_gate |
| Destructive resource operations require a preceding gate | Before every step | critical | step_requires_gate |
| External content taint: high-impact actions require a fresh gate | Before every step | critical | not |
| Bound resource calls per task to prevent runaway loops | Before every step | high | execution_max_steps |
Version 1.0.2. Source: manifests/security/owasp-agentic-default.yaml.
Assign it
One command, no redeploy.
Assign it from the dashboard or the CLI. Running agents pick up an assignment within the policy refresh window, five minutes by default.
- Policies
- 8, of which 5 are checked before every step.
- Version
- 1.0.2
- OWASP agentic
- ASI01, ASI02, ASI04, ASI05, ASI06, ASI08, ASI09, ASI10
$ kyvvu list-manifests$ kyvvu assign-manifest --agent-id <agent> --repo-id <repo> \ --manifest manifests/security/owasp-agentic-default.yamlAbout this pack
This template implements the structural minimum for safe agent operation: - Explicit, version-controlled goals and tool allowlists (registration) - Mandatory gates on irreversible and high-impact actions (step execution) - Path-dependent enforcement when external content has entered the task - Bounds on runaway behaviour
Pair with a compliance template (e.g. EU AI Act) for regulatory coverage.
Runtime policies in this pack
- Tool calls must be in the declared allowlist (critical)
- Code execution requires a preceding gate (critical)
- Destructive resource operations require a preceding gate (critical)
- External content taint: high-impact actions require a fresh gate (critical)
- Bound resource calls per task to prevent runaway loops (high)
OWASP agentic threats it addresses
- ASI01: Agent Goal Hijack
- ASI02: Tool Misuse and Exploitation
- ASI04: Agentic Supply Chain Vulnerabilities
- ASI05: Unexpected Code Execution (RCE)
- ASI06: Memory & Context Poisoning
- ASI08: Cascading Failures
- ASI09: Human-Agent Trust Exploitation
- ASI10: Rogue Agents
From Kyvvu's open policy library at github.com/Kyvvu/manifests, Apache 2.0.
Next
Run the whole pack in your browser.
In your browser
Try it in the playground
The policy is loaded. Add the steps your agent takes and watch the engine decide, step by step.
On your machine
Five minutes, no account
kyvvu try runs a demo agent against the bundled baseline. Then assign this pack from the dashboard or the CLI.
For security teams
See it on your agents
A 30-minute demo on the agents you actually run, with the policies that fit them.