git-hub-issus

ReviewAudited by ClawScan on May 1, 2026.

Overview

This instruction-only Git workflow helper is aligned with its purpose, but users should review commands because it can change and push code using their configured repository access.

This skill appears safe to install as an instruction-only Git helper if you intend the agent to assist with branch, commit, pull, push, and PR/MR workflows. Before using it, make sure the repository and remote are correct, review the diff, and require explicit confirmation for commits, pushes, branch deletion, rebases, or any force/reset operation.

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.

Low

#ASI02: Tool Misuse and Exploitation

What this means

If used without review, unintended local changes could be committed or pushed to a shared remote branch.

Why it was flagged

These are purpose-aligned Git workflow examples, but they can stage all changes, create commits, and push code to a remote repository.

Skill content

git add -A
git commit -m "feat: add dark mode support (closes #1)"
git push -u origin feature/issue-1-add-dark-mode

Recommendation

Review `git status` and `git diff`, confirm the target branch and remote, and get explicit user confirmation before commit or push actions.

Medium

#ASI03: Identity and Privilege Abuse

What this means

The agent could create repository changes or PR/MR records under the user's account if the user authorizes those workflows.

Why it was flagged

Creating pull or merge requests and pushing branches normally uses the user's configured GitHub, GitLab, SSH, or Git credentials. This is expected for the skill's purpose, but it is delegated account authority.

Skill content

- GitHub: use `github` skill → `gh pr create`
- GitLab: use `gitlab-ops` skill → `glab mr create`

Recommendation

Use appropriately scoped GitHub/GitLab credentials and verify the repository, organization, and visibility before creating PRs/MRs or pushing branches.

Info

#ASI04: Agentic Supply Chain Vulnerabilities

What this means

Users have less context about who authored or maintains the instructions, though there is no executable payload shown.

Why it was flagged

The skill has limited provenance information, but it also has no executable code or install steps in the provided artifacts.

Skill content

Source: unknown
Homepage: none
No install spec — this is an instruction-only skill.

Recommendation

Inspect the instruction text before enabling it and prefer skills from sources you trust.