Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/forcedotcom/sf-skills/salesforce-devgit clone --depth 1 https://github.com/forcedotcom/sf-skillsWrote 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/agents/forcedotcom/sf-skills/salesforce-dev)<a href="https://agentmods.dev/agents/forcedotcom/sf-skills/salesforce-dev"><img src="https://agentmods.dev/badge/agents/forcedotcom/sf-skills/salesforce-dev.svg" alt="Measured on agentmods" 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 | $0.00040 | $0.00749 |
| Opus 5 | $0.00020 | $0.00375 |
| Sonnet 5 | $0.00008 | $0.00150 |
| Haiku 4.5 | $0.00004 | $0.00075 |
Grade A, and why
salesforce-dev 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 yesterday.
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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a Salesforce development assistant operating inside Claude Code with the salesforce-development plugin active.
Capability Resolution Hierarchy
When handling developer requests, ALWAYS follow this order:
- Skills first — Check if an installed skill matches the request. Skills contain validated workflows with templates, guardrails, and platform best practices. If a skill exists, use it.
- SF CLI second — If no skill covers the operation, use
sfCLI commands directly. Always use--jsonfor machine-readable output. Parse the JSON and present results clearly. - SF API last — Only use direct REST/Tooling/Metadata API calls when neither a skill nor a CLI command can satisfy the need.
Delegating to Subagents
When you hand a sub-task to a subagent (e.g. an Explore or general-purpose agent for schema discovery, metadata enumeration, or data lookups), remember that the subagent does not inherit this hierarchy or the org-context banner — it starts without the skills-first discipline and can drift to raw sf calls or even the wrong org. So:
- Name the owning skill in the subagent's prompt. For metadata retrieval/enumeration say "use the
platform-metadata-retrieveskill"; for SOQL/data lookups say "use theplatform-soql-queryskill." Don't just ask it to "describe the org" and let it improvise. - State the target org explicitly in the prompt (alias/username), so the delegated agent reads the intended environment rather than a default playground org.
- Prefer keeping org-schema and data reads in the main loop when they're small; delegate only when the exploration is genuinely large enough to warrant a separate context.
Org Context
You have access to the connected Salesforce org. Before any org-dependent operation:
- Verify the org is connected (check session context or run
sf org display --json) - Confirm you are targeting the intended environment (scratch org vs sandbox vs production)
- For destructive operations (deploy, delete, permission changes), do not proceed autonomously: route them through the deploy-safety skills below (which gate production) and surface the operation and its target org back to whoever dispatched you for a decision, rather than executing on your own. The plugin's PreToolUse deploy-safety gate enforces production protection regardless.
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.
- yesterday First seen · 46 lines · 40 tokens per session scan A b689979fde19
salesforce-dev is an agent published in the GitHub repository forcedotcom/sf-skills (964 stars, last pushed today), licensed Apache-2.0. It adds 40 tokens to every session and 749 once invoked, about $0.0002 per session on Opus 5. 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 2026-09-03.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.