Sharpagent Engineering Lifecycle

PassAudited by ClawScan on May 11, 2026.

Overview

This is an instruction-only engineering workflow skill that mainly guides the agent to plan, edit, test, and review code; its file-changing authority is expected for that purpose.

This looks safe to use as a structured coding workflow, but it is designed to let the agent edit project files. Run it in a repository with backups or version control, review generated specs and code diffs, and do not treat the skill’s self-labeled “verified” status as independent validation.

Findings (2)

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

Invoking the skill may lead the agent to create or edit specs, tasks, tests, and code in the current project.

Why it was flagged

The skill expects authority to read and change project files. That is normal for an engineering lifecycle skill, but it is still user-impacting because it can modify local code and documents.

Skill content
preconditions:\n    - "Access to file system for reading/writing code"
Recommendation

Use it in a version-controlled workspace and review diffs before accepting or shipping changes.

What this means

A user might over-trust the workflow because of the wording, even though the supplied registry information lists no homepage or source repository.

Why it was flagged

The skill self-describes its trust level as verified. This appears to be a self-attestation inside the skill text, not independent proof of provenance or safety.

Skill content
trust_level: verified
Recommendation

Treat the workflow as unverified unless you have independent provenance or have reviewed the instructions yourself.