Skillv1.0.0

ClawScan security

Linkedin Content Planner · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 14, 2026, 9:40 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is an instruction-only planner for LinkedIn content with no code, no installs, and no requested credentials — its requirements and behavior align with the stated purpose.
Guidance: This skill appears coherent and low-risk: it only contains human-readable instructions for planning LinkedIn content. Before using it, keep these practical precautions in mind: (1) Do not hand over LinkedIn account credentials, access tokens, or other secrets to this skill unless you trust the publisher — this skill does not request them, but related 'executor' skills (e.g., post-writer or comment-drafter) might. (2) If you provide 'voice sample' file paths or URLs, confirm they contain only content you intend to share; the skill may use them to mimic tone. (3) The planner advises operational tactics (timed comments, DMs, early commenting) that can have ethical or policy implications — apply your own judgment and LinkedIn's terms of service. (4) If you later install or invoke the related automation skills that post or message on your behalf, review their install specs and env requirements carefully for credential or network access.

Purpose & Capability: okName/description match the SKILL.md: the files only contain guidance for producing a 7-day LinkedIn content plan and related checklists. There are no unrelated environment variables, binaries, or install steps requested.
Instruction Scope: noteThe instructions stay within content-planning scope (pillar mix, formats, hooks, comment targets, posting times). Two small notes: (1) 'Voice samples (optional): paths to past posts for voice calibration' implies the agent may be given or asked to read user-provided files/URLs — the skill does not itself declare or require file access, but an operator could supply paths; (2) the doc recommends active behaviors (commenting, DMs, timing strategies) which are operational/social tactics rather than system-level actions. The SKILL.md does not instruct reading system env vars, secrets, or contacting external endpoints directly.
Install Mechanism: okNo install spec and no code files present — instruction-only. This minimizes disk writes and code execution risk.
Credentials: okThe skill requests no environment variables, credentials, or config paths. That is proportionate for a planner that only generates content recommendations.
Persistence & Privilege: okalways is false and the skill does not request permanent presence or to modify other skills or system settings. Autonomous invocation is allowed by platform default but this skill itself has no elevated privileges.