ClawHub

Linkedin Writer 1.0.0

v1.0.0

Writes LinkedIn posts that sound like a real person, not a content mill

⭐ 1· 571·2 current·2 all-time
by@rezcarbon
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 SKILL.md instructions (guidelines for writing LinkedIn posts). No unrelated binaries, env vars, or access are requested. Minor provenance inconsistency: registry Owner ID (kn7b3j3...) differs from _meta.json ownerId (kn76xhdy...), and source/homepage are missing — this is a metadata mismatch but does not change the skill's functional coherence.
✓
Instruction Scope
SKILL.md contains only writing templates, formatting/voice rules, and questions to ask the user for context. It does not instruct the agent to read files, access environment variables, call external endpoints, or transmit data outside the normal agent flow.
✓
Install Mechanism
No install spec and no code files are present — instruction-only skills have minimal on-disk/supply-chain risk.
✓
Credentials
No environment variables, credentials, or config paths are requested; the skill only asks for user-provided topic/story details as expected for a writing assistant.
✓
Persistence & Privilege
always is false and the skill does not request persistent presence or elevated privileges. It does not modify other skills or system settings.
Assessment
This skill is instruction-only and internally consistent with its purpose. Before using: (1) don't paste sensitive or confidential information (client secrets, full client names, private messages) into prompts — the skill asks for story details and could include them verbatim in output; (2) test generated posts in a safe draft before posting publicly to ensure tone and factual accuracy; (3) note the metadata mismatch (different ownerId in _meta.json vs registry) and missing homepage/source — if provenance matters to you, ask the publisher for confirmation before trusting or widely deploying the skill; (4) README links to external context packs are just references — the skill does not automatically fetch them, but if you or the agent follow those links, treat them like any external site and verify content before using.

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

latestvk97e7fypd4qh0jpb4nc37v680181qv4g

License

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

Comments