Skillv1.0.0

ClawScan security

Batch Processing (DataLoader) · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 4, 2026, 5:57 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is a self-contained JavaScript DataLoader implementation whose code, docs, and runtime instructions align with its stated purpose; there are no unexplained environment or network requirements.
Guidance: This package appears coherent and safe for evaluation. Before using in production: 1) Review and test the batchLoadFn implementations you supply (ensure they use parameterized queries to avoid SQL injection and actually batch work rather than making per-id calls). 2) Note the DataLoader caches Promises immediately — if a batch promise rejects, the rejected Promise remains cached (consider clearing cache on error). 3) Follow the SKILL.md best practice to create loaders per request to avoid cache pollution. 4) Run the provided examples/benchmarks in a sandboxed environment first. 5) If you need provenance, confirm the repository/source (package.json repo URL is present but source was listed as unknown).

Purpose & Capability: okName, description, SKILL.md, package.json and the three JS files all consistently implement a DataLoader pattern for batching and caching; nothing in the bundle asks for unrelated services, credentials, or system access.
Instruction Scope: noteSKILL.md and examples focus on creating loaders that call db.query or fetch; these are example usages rather than instructions to read system files or secrets. Note: example code references network URLs and a fetch-like API only in examples (mocked in examples.js) — the core implementation itself does not perform external network I/O.
Install Mechanism: okNo install spec is present (instruction-only install), and the package is pure JS with zero external dependencies. Nothing is downloaded from arbitrary URLs or written to system paths as part of installation.
Credentials: okThe skill declares no required environment variables, no credentials, and the code does not access process.env or system config paths. Requested privileges are minimal and appropriate for a library.
Persistence & Privilege: okFlags such as always and disable-model-invocation are default/normal. The skill does not request persistent agent privileges or attempt to modify other skills or global configuration.