Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Corall
v0.9.0Handle the Corall marketplace — setup, order handling, and order creation. Triggers when: (1) a hook message has Task name "Corall" or session key contains "...
⭐ 0· 35·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill is about operating the Corall marketplace via the corall CLI and all declared requirements list only the corall binary — that matches the stated purpose. However the references and SKILL.md also rely on other local tools and services (openclaw, curl, jq, python3) and touch OpenClaw config and Corall credential files; these extra capabilities are plausible for this integration but are not fully declared in the metadata.
Instruction Scope
Runtime instructions read and (via the corall CLI) write local config/credential files (~/.corall/*.json and ~/.openclaw/openclaw.json), call external services (api.corall.ai, Stripe checkout links, R2 presigned uploads), and may perform network checks (curl to api.ipify.org). The SKILL.md includes explicit safeguards (do not upload pre-existing host files in webhook mode; confirm artifact uploads with the user) but also instructs actions that can reveal or modify local credentials/configs. These behaviors are within the marketplace integration scope but elevate risk and should be confirmed with the user before execution.
Install Mechanism
This is an instruction-only skill with no install spec (lowest install risk). The skill recommends running the corall CLI's own upgrade/install (which fetches releases and replaces the binary in-place) and points users to corall.ai; that external fetch is not performed by the skill itself but is a recommended operator action and thus a user decision. Verify the source before running upgrades.
Credentials
No environment variables or primary credentials are requested by the skill metadata (good). However the instructions access local credential files (~/.corall/credentials/*.json) and the OpenClaw config; these are necessary for the Corall integration but constitute access to sensitive secrets/config. Also helper tools used in examples (openclaw, curl, jq, python3) are not declared in required binaries, which is an inconsistency the integrator should address.
Persistence & Privilege
The skill does not request always:true and does not autonomously persist itself. It does instruct running corall openclaw setup, which merges Corall settings into the OpenClaw config (modifies ~/.openclaw/openclaw.json and adds 'hook:' to allowedSessionKeyPrefixes and gateway settings). Modifying host OpenClaw config is expected for webhook integration but is a privileged action — review changes before applying.
What to consider before installing
This skill is coherent with a Corall CLI integration, but review these before you use it: 1) The skill expects to run corall commands that will read/write ~/.corall credentials and may modify your OpenClaw config (~/.openclaw/openclaw.json). Those are sensitive files—confirm and back them up before running any setup commands. 2) The docs call external endpoints (api.corall.ai, Stripe checkout URLs, and presigned R2 upload URLs); artifact uploads send data off-host — never upload pre-existing host files in webhook mode and always confirm uploads with the user. 3) Examples use other tools (openclaw, curl, jq, python3) not declared in the skill metadata; ensure those binaries are available and safe. 4) The skill recommends running corall upgrade or installing from corall.ai (external fetch that replaces the binary) — verify the download source and checksum before upgrading. If you need lower privilege, require explicit confirmation for any config writes or uploads and avoid running upgrade/install steps automatically.Like a lobster shell, security has layers — review code before you run it.
latestvk97br53t2rs7sebz6c5gkr4k3184s70g
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🪸 Clawdis
Binscorall