Praesidia
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Anyone installing it must provide a Praesidia API key, which may allow access to organization and agent security configuration depending on the key's scope.
The skill requires a Praesidia API credential, and its examples use it as a bearer token for account-backed API requests.
metadata: {"openclaw":{"requires":{"env":["PRAESIDIA_API_KEY"]},"primaryEnv":"PRAESIDIA_API_KEY"}}Use a least-privilege Praesidia API key if available, store it securely, and rotate or revoke it if no longer needed.
A user-directed action can change how an agent is moderated or secured in Praesidia.
The skill documents a POST request that creates or enables an organization guardrail for a specific agent.
web_fetch({ url: "${PRAESIDIA_API_URL}/organizations/${orgId}/guardrails", method: "POST" ... isEnabled: true })Before allowing guardrail changes, confirm the organization ID, agent ID, template, action, scope, and whether the change should be enabled immediately.
Users may over-rely on a trust score as proof that an agent is completely safe.
The example wording could be read as a broad safety guarantee based on a verification status and trust score.
✅ ChatBot V2 is verified and safe to use!
Present verification results as Praesidia's reported status and score, and avoid implying that any score is an absolute safety guarantee.