Skillv1.0.0

ClawScan security

Project Context Generator For AI · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 16, 2026, 8:10 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code, templates, and runtime instructions are coherent with its stated purpose of generating a .ai-context knowledge base; no evidence of credential requests, network exfiltration, or unrelated capabilities was found — but note a minor runtime mismatch (the included script expects the Bun runtime while the skill metadata declares no required binaries).
Guidance: This skill appears to be what it says: a generator for a .ai-context knowledge base. Before installing or running it, do the following: - Inspect scripts/generate.ts yourself (it is included) and run it in a sandbox or review its behavior; it uses only fs/path and template rendering in the copy provided. - Ensure you have the runtime required to run the script (the script uses a Bun shebang). Either install Bun or run the script via an appropriate Node/TS workflow; the registry metadata should have declared this but does not. - Give the agent access only to the project workspace you intend it to read. The skill reads local project files (README.md, package.json, AGENTS.md, docs/, source tree); avoid granting repository-wide or system-wide file access if not needed. - Confirm there are no project-local templates or meta/scripts that call out to remote URLs or perform network operations before letting an agent run automated generation. - If you want tighter operational clarity, ask the publisher to update the metadata to declare the required runtime (bun/node) and to document any network or side-effecting behavior; absence of those declarations is likely an oversight, not malicious intent.

Review Dimensions

Purpose & Capability: noteThe skill's name, README, SKILL.md, templates, and scripts all align: they generate a .ai-context directory and read project files (README.md, package.json, AGENTS.md, docs/, source tree) to produce documentation. One inconsistency: the shipped script (scripts/generate.ts) uses a Bun shebang and TypeScript, so a Bun/Node runtime is effectively required to run the automation — but the skill's registry metadata lists no required binaries. This is likely an omission rather than malicious, but it is a capability/requirement mismatch the user should be aware of.
Instruction Scope: okSKILL.md explicitly instructs the agent to read local project files (AGENTS.md, README.md, package.json, docs/, source tree) and to generate docs from templates. All referenced actions are within the project scope and relevant to creating a knowledge base. There are no instructions to read unrelated system files, environment variables, or to transmit data to external endpoints.
Install Mechanism: noteNo install spec is provided (instruction-only), which minimizes risk. However, the included script is executable (#!/usr/bin/env bun) and expects Bun to run; the absence of a declared runtime/install step or required-binaries entry is an operational gap. There are no downloads or external install URLs in the files reviewed.
Credentials: okThe skill does not request any environment variables, credentials, or config paths. The files and templates likewise do not reference secrets or external service tokens. Requested file reads are limited to project-local files and templates, which is proportionate to the described function.
Persistence & Privilege: okThe skill is not configured as always-on and does not request persistent system privileges or attempt to modify other skills or global agent settings. It creates/updates files under a project-local .ai-context directory, which is appropriate for its purpose.