Praesidia
v0.1.0Verify AI agents, check trust scores (0-100), fetch A2A agent cards, discover marketplace agents, apply guardrails for security and compliance. Use when user mentions agent verification, trust scores, agent discovery, A2A protocol, agent identity, agent marketplace, guardrails, security policies, content moderation, or asks "is this agent safe?" or "find agents that can [task]" or "apply guardrails to protect my agent".
⭐ 0· 1.3k·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill name and description describe agent verification, trust scores, discovery, and guardrail management. The only required credential is PRAESIDIA_API_KEY and the SKILL.md shows calls exclusively to the praesidia API (agents/, organizations/, guardrails/). These requirements are proportionate and expected for the stated functionality.
Instruction Scope
All runtime instructions are web_fetch calls to Praesidia endpoints (GET agent-card, discovery, guardrails templates, POST guardrails, validate). The SKILL.md instructs the agent to obtain an orgId from 'profile or context' (not declared as an env variable) — this is plausible (the key can be used to query a profile endpoint) but means the skill expects to read contextual profile data or call a profile API. Also, the skill contains instructions to create/modify guardrails (POST) which will change org configuration when invoked with a key that has write/admin privileges.
Install Mechanism
Instruction-only skill with no install spec and no code files. No downloads or package installs are requested, so the platform risk from installation is minimal (nothing is written to disk by the skill itself).
Credentials
The skill requests only PRAESIDIA_API_KEY (declared as primaryEnv). This aligns with API usage. However, because the skill's actions include POSTing guardrails and modifying org-level policies, users should ensure the provided API key has appropriate, minimal scopes (e.g., read-only vs admin) and understand that a full-permission key would allow changes to guardrails and org state.
Persistence & Privilege
always:false (no forced always-on). The skill is allowed to be invoked autonomously by the agent (disable-model-invocation:false), which is platform default. Given the skill can modify guardrails via API calls, autonomous invocation combined with a high-privilege API key could result in policy changes without explicit human confirmation — this is expected for a management-style skill but worth considering when granting the key.
Assessment
This skill appears internally consistent with its purpose. Before installing, verify the Praesidia service and that you trust the skill source. Provide an API key with the least privileges necessary (prefer a read-only key if you only need verification/discovery). If you must allow guardrail modifications, use a scoped admin key and enable audit/logging on the Praesidia side. Confirm how orgId is resolved (profile API vs external config) and ensure the key cannot be used to alter unrelated org settings. Finally, be aware that if the agent invokes the skill autonomously it can make POST requests that change guardrails — restrict invocation or keys if you require human approval for policy changes.Like a lobster shell, security has layers — review code before you run it.
latestvk974ryy3jev710y3khpgpbba3h80k48c
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🛡️ Clawdis
EnvPRAESIDIA_API_KEY
Primary envPRAESIDIA_API_KEY