Ghostmeet

Security checks across static analysis, malware telemetry, and agentic risk

Overview

Ghostmeet’s instructions match its meeting-transcription purpose, but installing it requires external Docker and Chrome-extension code and it handles sensitive meeting transcripts.

Install only if you are comfortable running the external Ghostmeet backend and Chrome extension. Audit or pin the repository first, protect the Anthropic API key, confirm meeting participants are comfortable with transcription, and avoid sending sensitive transcripts for AI summary unless that is acceptable.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

The installed backend or extension could access meeting audio or transcripts, so the safety of those external files matters.

Why it was flagged

The setup pulls and runs external backend and extension code that is not part of the reviewed skill artifact and is not pinned to a commit.

Skill content
git clone https://github.com/Higangssh/ghostmeet.git ... docker compose up -d ... Chrome Extension must be installed in developer mode from `extension/` folder.
Recommendation

Review or pin the Ghostmeet repository before installing, audit the Chrome extension permissions, and run it only in an environment you trust.

What this means

A local Ghostmeet service may remain available on your machine until you stop it.

Why it was flagged

The backend is started in detached mode, so it can continue running after the immediate chat task is complete.

Skill content
docker compose up -d
Recommendation

Stop the backend when finished, for example with Docker Compose from the Ghostmeet directory, and disable or remove the browser extension when not in use.

What this means

If the key is mishandled, someone else could use your Anthropic account or incur API costs.

Why it was flagged

Summary generation uses an Anthropic API credential, which is expected for the feature but still needs normal secret-handling care.

Skill content
`GHOSTMEET_ANTHROPIC_KEY` — required for AI summary generation.
Recommendation

Use a dedicated API key if possible, keep the `.env` file private, monitor usage, and revoke the key if you no longer need it.

What this means

Private meeting details may be displayed in the current chat or used to answer later questions about sessions.

Why it was flagged

The skill can retrieve stored meeting transcripts, which are sensitive contextual records of private conversations.

Skill content
Fetch transcripts → retrieve full text from a session
Recommendation

Ask for specific sessions when possible, avoid recording highly sensitive meetings unless appropriate, and delete old transcripts according to your privacy needs.

What this means

Meeting transcript content leaves the local machine when AI summaries are generated.

Why it was flagged

The artifact clearly discloses that generating summaries sends transcript text to an external AI provider.

Skill content
Summaries use Anthropic API — when you click Summarize, transcript text is sent to Claude API.
Recommendation

Generate summaries only with participant consent and only for meetings whose contents you are comfortable sending to the configured AI provider.