Saleor Commerce
PassAudited by ClawScan on May 13, 2026.
Overview
The visible artifacts look like a Saleor development guidance skill with local safety checks; the main cautions are expected web browsing for docs and powerful Saleor app permissions if you use it to build real integrations.
This appears safe to install as a Saleor development aid. Before using it on a real store, verify generated code and manifests, keep Saleor tokens and payment credentials server-side, limit app permissions, and review any database migration, deployment, or destructive command before running it.
Findings (4)
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.
The agent may browse Saleor and related developer documentation before producing implementation code.
The skill directs the agent to use web/search tools before coding. This is disclosed and aligned with keeping Saleor API usage current, but it does force external lookup behavior.
Before writing any Saleor implementation code, you MUST web-search and/or web-fetch the relevant official documentation.
Use the skill when web access is acceptable, and instruct the agent explicitly if you need offline-only work or want specific sources used.
If this hook is enabled, some database, Docker, migration, or deployment commands may be warned about or blocked until reviewed.
The helper script inspects Bash commands and blocks or warns on destructive Saleor/Django/database operations. This is protective and scoped to the development purpose.
PreToolUse hook: block potentially destructive Saleor / Django commands ... python\s+manage\.py\s+flush ... DROP\s+(TABLE|DATABASE) ... sys.exit(2)
Treat the hook as a useful guardrail, but still manually review production database, deployment, and destructive filesystem commands.
If enabled, generated file contents may be locally scanned for likely secrets and the agent may receive a warning.
The helper script inspects content being written or edited and emits a local warning for hardcoded secret patterns. The provided code does not transmit or persist the content.
if tool_name == "Write": content = tool_input.get("content", "") ... Security notice: Possible hardcoded secret(s) detectedUse this as a supplemental safeguard, but continue to keep real secrets in environment variables or a secrets manager.
Code or app manifests generated from this guidance could request permissions that allow broad changes to a Saleor store if installed with those scopes.
Saleor Apps legitimately use stored app tokens and permissions, including permissions that can manage store data, users, apps, and payments. The skill also advises least privilege.
The APL stores the mapping between Saleor instance URLs and the App's auth tokens ... MANAGE_PRODUCTS ... MANAGE_USERS ... MANAGE_APPS ... HANDLE_PAYMENTS
Review every generated Saleor App manifest, request only required permissions, store tokens server-side, and avoid using production credentials during experimentation.