ClawHub

Core Message Extractor

v1.0.0

Extract the single-sentence core of any message — the one thing that must survive if everything else is lost. Use this skill whenever the user asks 'what's m...

⭐ 0· 31·0 current·0 all-time
byHung Quoc To@quochungto
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
CryptoCan make purchases
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name, description, and tasks (distill drafts to a single-sentence core) align with the SKILL.md content. The skill only needs short-form text, audience, and goal to function — nothing requested appears unrelated to messaging work.
ℹ
Instruction Scope
Runtime instructions are prose-only and focus on extracting a core sentence. The skill instructs the agent to look for draft files (e.g., draft.md, brief.md, README.md, core-message.md) and optionally check git history for message files; this is reasonable for a writing/messaging skill but means the agent will examine workspace files and repo history if present. There are no instructions to transmit data to external endpoints or to access unrelated secrets.
✓
Install Mechanism
No install spec and no bundled code files — lowest-risk installation surface. The skill is instruction-only and will not write new binaries to disk.
✓
Credentials
The skill requests no environment variables, credentials, or config paths. Suggested file checks (drafts, brand docs, prior cores) are proportional to its stated purpose; nothing asks for unrelated credentials or secret data.
✓
Persistence & Privilege
always is false and the skill is user-invocable; it does not request permanent elevated privileges or attempt to modify other skills or system-wide settings.
Assessment
This skill is instruction-only and appears coherent for messaging work. Before installing, consider: it will look at files in the agent's working directory (draft.md, brief.md, README.md, core-message.md, and possibly git history) to gather context — avoid invoking it in workspaces containing sensitive or unrelated documents you don't want the agent to read. There are no network endpoints, installs, or credential requests in the SKILL.md. If you want maximum safety, review the SKILL.md yourself and run the skill only on the specific draft text you provide rather than granting broad repository access.

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

bookforgevk972mzgg8d4w0xmj2njq3csxss84wewhcommanders-intentvk972mzgg8d4w0xmj2njq3csxss84wewhcommunicationvk972mzgg8d4w0xmj2njq3csxss84wewhcopywritingvk972mzgg8d4w0xmj2njq3csxss84wewhlatestvk972mzgg8d4w0xmj2njq3csxss84wewhmessagingvk972mzgg8d4w0xmj2njq3csxss84wewhpitchingvk972mzgg8d4w0xmj2njq3csxss84wewhprioritizationvk972mzgg8d4w0xmj2njq3csxss84wewh

License

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

Runtime requirements

🎯 Clawdis

Comments