ClawHub

Intercom

v1.0.1

Create, update, improve, and review Intercom help-center and support documentation. Use when writing new Intercom articles, revising existing docs, auditing...

⭐ 0· 57·0 current·0 all-time
byGeorge Lewis@georgelewi5
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
The skill's name and description match the SKILL.md: it focuses on drafting, reviewing, and improving Intercom help-center content. The only capability beyond pure writing is optional Intercom API interaction, which is appropriate for a tool that can inspect or update workspace content.
✓
Instruction Scope
SKILL.md stays on-topic: it prescribes writing/review workflows and only references the Intercom API when a private workspace access token is available. It explicitly recommends read-only inspection first, showing changes before applying them, and warns not to write tokens into docs/commits/skill files.
✓
Install Mechanism
No install spec or code files are present (instruction-only), so nothing is written to disk and no external packages are pulled in.
✓
Credentials
The skill does not require any environment variables, binaries, or config paths. The only external credential discussed is an Intercom workspace access token, and the doc treats it as optional and sensitive—appropriate and proportional for the described API interactions.
✓
Persistence & Privilege
always is false and the skill does not request persistent system presence or elevated privileges. It also instructs caution around token handling and prefers showing deltas before applying changes.
Assessment
This skill is coherent and instruction-only. Before installing or using it: (1) do not supply an Intercom access token unless you trust the agent and want live workspace inspection/updates; (2) if you must provide a token, prefer a least-privilege/read-only token and monitor Intercom API logs; (3) require the agent to show proposed edits and get explicit approval before applying changes (the SKILL.md already recommends this); and (4) avoid storing tokens in commits, docs, or skill files. If you need tighter controls, use an OAuth app or a temporary scoped token for one-off reviews.

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

latestvk976xaxwszrfda144j2efbk6d183qn4f

License

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

Comments