Back to skill
Skillv0.1.0
ClawScan security
Communication Playbook · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 9, 2026, 12:16 AM
- Verdict
- benign
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill is an instruction-only communication playbook that asks for no credentials, performs no I/O, and its requested capabilities align with its stated purpose — although its source is unknown and its broad trigger language may lead to frequent autonomous use.
- Guidance
- This skill is an instruction-only playbook and appears internally consistent and low-risk. Before installing or enabling it: 1) verify the skill's origin (registry shows unknown source and README references a GitHub repo); don't run any npx or install commands until you inspect that repository. 2) Review SKILL.md to ensure the communication guidance matches your legal/regulatory needs (the playbook advises consulting legal for regulator responses). 3) If you are concerned about frequent autonomous use, restrict when the agent can auto-invoke the skill (or require explicit user invocation) so the broad 'when in doubt, use it' trigger doesn't cause noisy or unintended changes. 4) For enterprise use, have an admin or security reviewer approve the skill and its triggers before enabling it broadly.
Review Dimensions
- Purpose & Capability
- okName and description match the content of SKILL.md: a playbook for internal/external communications and related artifacts. The skill requests no binaries, env vars, or config paths, which is proportionate for a guidance-only playbook.
- Instruction Scope
- noteSKILL.md contains only prescriptive guidance for writing/reviewing communications and lists when to trigger the skill. It does not instruct the agent to read files, use credentials, call external endpoints, or collect system data. One operational concern: the instructions use very broad trigger language (e.g., 'When in doubt, use it'), which may cause the agent to apply this playbook in many contexts.
- Install Mechanism
- okThere is no install spec and no code files; this is instruction-only, which is lowest-risk. The README contains an example npx install command pointing at a GitHub repo, but that is only documentation in README.md — the skill itself includes no install instructions in the registry metadata. If you plan to run any install command from the README, verify the repository and contents before executing.
- Credentials
- okThe skill requires no environment variables, no credentials, and no config paths. There is no evidence of excessive credential requests or access needs inconsistent with a playbook.
- Persistence & Privilege
- notealways is false (normal). The platform default allows autonomous invocation; combined with the playbook's broad trigger rules, the skill could be invoked frequently by agents without explicit user prompting. This is a behavior characteristic to monitor but not an inherent security flaw in the skill content.