Borrowing it
Nothing to install: this file belongs to ehebert7/salesforce-claude-framework. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ehebert7/salesforce-claude-framework/master/.claude/commands/setup.mdgit clone --depth 1 https://github.com/ehebert7/salesforce-claude-frameworkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/ehebert7/salesforce-claude-framework/setup)<a href="https://agentmods.dev/commands/ehebert7/salesforce-claude-framework/setup"><img src="https://agentmods.dev/badge/commands/ehebert7/salesforce-claude-framework/setup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/ehebert7/salesforce-claude-framework/setup"><img src="https://agentmods.dev/badge/commands/ehebert7/salesforce-claude-framework/setup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01180 |
| Opus 5.5 | $0.00000 | $0.00472 |
| Sonnet 5.5 | $0.00000 | $0.00236 |
| Haiku 4.5 | $0.00000 | $0.00118 |
Grade A, and why
setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Framework Setup
Configure the Salesforce Claude Code Framework for your environment.
Note: If the framework isn't installed yet, use /install first.
Process
Step 1: Verify Installation
Check if framework files exist:
.claude/hooks/skill-rules.json.claude/commands/dev-docs.md
If missing, display:
Framework not found in current directory.
Run /install first to install the framework, then run /setup to configure it.
And stop.
If found, continue.
Step 2: Skill Configuration
Display:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
SKILL CONFIGURATION
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Skills are markdown files containing domain knowledge that Claude
can reference when helping you code.
The framework can auto-detect keywords in your prompts and suggest
relevant skills.
Ask:
Do you have skills you'd like to integrate?
1. Scan a directory for skills
2. Use skills from my home directory (~/.claude/skills/)
3. Skip - I'll configure manually later
4. Show me how skills work
If Option 1 (Scan directory):
Ask: "Enter the path to your skills directory:"
Scan the directory for .md files recursively. For each file:
- Read the filename (without .md) as skill name
- Optionally read the first heading as display name
Display found skills:
Found X skills:
• apex-best-practices
• lwc-dev-guidelines
• agentforce-setup
• salesforce-flow-architect
[...]
Would you like to configure auto-detection triggers for these skills?
1. Yes - configure each one
2. Yes - use smart defaults based on skill names
3. No - just register them without triggers
If "smart defaults": Generate triggers based on skill name:
apex-best-practices→ keywords: ["apex", "trigger", "class", "soql", "dml"]lwc-dev-guidelines→ keywords: ["lwc", "lightning", "component", "wire"]agentforce-*→ keywords: ["agentforce", "agent", "bot", "topic", "action"]- etc.
If "configure each": For each skill, ask:
Skill: [skill-name]
Enter keywords (comma-separated) that should trigger this skill:
>
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 209 lines · 0 tokens per session scan A 320d5289d003
setup is a command published in the GitHub repository ehebert7/salesforce-claude-framework (12 stars, last pushed 11mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,180 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
API_REFERENCE
Generated from commands/specs/.json. Do not edit this file directly.
build-skills
End-to-end orchestrator that takes a count + optional topic hints and ships N production-grade skills with measurement-driven quality gates at every step.
onboard-source
One entrypoint for turning outside knowledge into catalog-quality skills.
clarify-requirements
Wraps agents/requirements-clarifier/AGENT.md. Takes one requirement — a paragraph, a worksheet, a backlog extract — searches local coverage for every capability it implies, harvests every row of every matching skill's ## Questions to Ask Before Configuring table, dedupes by meaning, proposes a default per question…
plan-build
Wraps agents/build-planner/AGENT.md. Takes a build directory whose plan.json is past gate G1 and writes the plan the rest of the loop executes: scope in and out, a fit-gap entry per capability, a decision per technology choice with the decision-tree branch that produced it, two to six milestones, and the ordered steps…
verify-plan
Wraps agents/plan-verifier/AGENT.md. For every step in plan.json it runs three independent adversarial lenses — executability (can the named agent actually produce these outputs from these inputs, and is it eligible to own the step in this buildmode?), grounding (do the cited skills, templates and decision-tree…