EPAI

ReviewAudited by ClawScan on May 10, 2026.

Overview

This is a coherent EPAI management CLI skill, but it uses EPAI credentials and can upload or delete knowledge-base content, so users should run destructive actions deliberately.

Install only if you intend to let the agent manage EPAI resources. Configure a least-privileged EPAI API key, verify API_BASE points to the correct EPAI service, and be especially careful with upload and delete commands.

Findings (3)

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

A mistaken delete command could remove EPAI knowledge bases, documents, or catalogs.

Why it was flagged

The skill clearly documents destructive EPAI operations such as deleting knowledge bases, catalogs, and documents. These operations are purpose-aligned and ID-scoped, but they can remove platform content if invoked with the wrong IDs.

Skill content
python scripts/epaiclt.py   --method kb_delete   --kb_ids kb_id1 kb_id2 ...
Recommendation

Confirm target IDs and keep backups or recovery plans before running delete operations.

What this means

Anyone or any agent process using these environment variables can perform the EPAI operations allowed by that key.

Why it was flagged

The CLI requires an EPAI API key and account identifier and sends them as request headers to the configured EPAI API base. This is expected for the integration, but it is still delegated account authority.

Skill content
API_KEY = os.getenv("EPAI_API_KEY")
ACCOUNT = os.getenv("EPAI_ACCOUNT")
HEADERS = {
    "Authorization": API_KEY,
    "Account": ACCOUNT
}
Recommendation

Use the least-privileged EPAI API key available, keep it out of logs and shared shells, and rotate it if exposed.

What this means

Sensitive local documents could become part of a persistent EPAI knowledge base if uploaded.

Why it was flagged

The document upload path reads selected local files and uploads them into an EPAI knowledge-base/document pipeline, where their contents may be parsed, embedded, and reused by the platform.

Skill content
url = f"{API_BASE}/document/upload"
upload_files = [("files", (os.path.basename(f), open(f, "rb"))) for f in files]
Recommendation

Upload only files intended for the target EPAI knowledge base and verify the destination kb_id before use.