salesforce — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited salesforce (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Claude Code and Cowork skill for Salesforce org interaction — authenticate, query, download and upload files.
Built-in safety: Write operations require user confirmation. Delete operations show explicit warnings and require typed confirmation. Read operations execute freely. Set SALESFORCE_SKIP_WARNINGS=true to bypass all confirmations.
In Claude Code, run:
/plugin marketplace add enzoleonardi/claude-salesforce-skill
/plugin install salesforce@enzoleonardi-claude-salesforce-skillmkdir -p ~/.claude/skills/salesforce
curl -sL https://raw.githubusercontent.com/enzoleonardi/claude-salesforce-skill/main/skills/salesforce/SKILL.md \
-o ~/.claude/skills/salesforce/SKILL.mdThis skill teaches Claude Code and Cowork how to:
| Category | Capabilities |
|---|---|
| Auth | Web login (OAuth), access token fallback, Cowork support |
| Read 🟢 | SOQL queries, bulk queries (10k+), Tooling API, object discovery |
| Write 🟡 | Create, update, upsert records, bulk upsert from CSV, data import |
| Delete 🔴 | Single record delete, bulk delete from CSV (with safety warnings) |
| Apex | Execute anonymous Apex, view/tail debug logs |
| Metadata | List types, retrieve components, deploy (with dry-run) |
| Files | Upload and download via ContentVersion, batch operations, multipart for large files |
| Monitor | API limits, org health, debug log inspection |
🟢 READ → Execute freely
🟡 WRITE → Warn user, ask confirmation
🔴 DELETE → Explicit warning, require typed confirmationTo bypass all warnings: export SALESFORCE_SKIP_WARNINGS=true
Once installed, Claude Code and Cowork will automatically use this skill when you mention Salesforce-related tasks:
> Query all closed-won opportunities from last year
> Create a new Account called "Acme Corp"
> Download all files attached to these accounts
> What custom objects exist in my org?
> Execute this Apex script against my org
> Show me my org's API limits
> Deploy this Apex class to my sandboxThe skill supports three methods (tried in order):
sf org login web opens the browser for OAuth. Secure, long-lived refresh token.claude-salesforce-skill/
├── .claude-plugin/
│ ├── plugin.json # Plugin manifest
│ └── marketplace.json # Marketplace definition
├── skills/
│ └── salesforce/
│ └── SKILL.md # The skill content
├── README.md
└── LICENSEMIT
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.