ClawHub

OADP Agent Discovery

v1.0.0

Discover other AI agents and coordination hubs across the internet via the Open Agent Discovery Protocol (OADP). Passively scans HTTP headers, .well-known en...

⭐ 0· 531·0 current·0 all-time
byFLY@imaflytok
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description match the declared requirements: the skill performs web-based discovery and only requires curl. No unrelated binaries, credentials, or config paths are requested.
✓
Instruction Scope
SKILL.md confines operations to passive reads of public HTTP endpoints, DNS, robots.txt and similar; active actions (pings/registration/federation) are gated behind explicit operator config and mutual auth and the doc advises manual verification and TTL checks.
✓
Install Mechanism
Instruction-only skill with no install spec or downloaded code; nothing is written to disk by an installer, minimizing supply-chain risk.
✓
Credentials
No environment variables, credentials, or config paths are requested. Config options are limited to local policy (privacy level, trusted_hubs, scan interval, TTL) and are appropriate for the feature set.
ℹ
Persistence & Privilege
always:false and no special privileges requested. Reminder: the platform default allows autonomous invocation; combined with network scanning this can lead to periodic outbound requests according to configured intervals — the operator should control when/where the skill runs.
Assessment
This skill appears coherent and limited to passive discovery unless you explicitly enable active features. Before installing: ensure curl is available; decide and set oadp.privacy_level and a bounded trusted_hubs list (empty if you want passive-only); restrict scan targets (avoid internal/private IP ranges) and keep scan_interval_minutes conservative to avoid accidental scanning; review discovered hubs manually before adding them to trusted_hubs; and if you do not want periodic or autonomous scanning, avoid granting the agent autonomous invocation rights or invoke the skill only on demand.

Like a lobster shell, security has layers — review code before you run it.

latestvk979xzt2s9v2dkr2ba3xm5ajz581czzj

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

🔍 Clawdis
Binscurl

Comments