Skip to main content

implementation-best-practices

Route and synthesize implementation decisions for Agentforce, Data 360, Slack, Tableau Next, Apex, APIs, prompts, MCP, sub-agents, state, long-running actions, testing, and user experience. Use after product and environment decisions are known and a partner needs a scalable implementation pattern.

Source facts

Repository
forcedotcom/partner-ecosystem-agentic-kit
Last source activity
November 4, 2026 at 20:04
Detected SKILL.md language
English
Stars
20
Forks
0

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer
3 files

Showing SKILL.md

SKILL.md
Source instructions · Read-only preview
name
implementation-best-practices
description
Route and synthesize implementation decisions for Agentforce, Data 360, Slack, Tableau Next, Apex, APIs, prompts, MCP, sub-agents, state, long-running actions, testing, and user experience. Use after product and environment decisions are known and a partner needs a scalable implementation pattern.
metadata
{"version":"1.0.0","contract-schema":"1.0.0","role":"capability-router","primary-capability":"implementation-best-practices","contract":"references/skill-contract.json"}
Follow the instructions below. Read [this skill's contract](references/skill-contract.json) only to clarify a handoff, missing dependency, or ambiguous authority boundary. It is a tooling declaration, not execution permission; the active primary owns synthesis. # Implementation Best Practices Advisor Own implementation-pattern synthesis while delegating product execution to dynamically discovered, approved Salesforce skill libraries. ## Authority modes - **Ask:** compare architecture and implementation patterns from supplied context only. - **Assess:** inspect approved designs, repositories, metadata, tests, and read-only org evidence; record what was actually inspected. - **Act:** present exact changes, tools, org authority, side effects, rollback, and validation; execute only after explicit approval. ## Focused routing - Cross-product pattern selection: `implementation-pattern-router`. - Headless Toolkit, MCP client/server direction, Salesforce hosted MCP tools, and HXL: `headless-mcp-advisor`. - Agentforce: `agentforce-implementation-advisor`. - Data 360: `data-360-implementation-advisor`. - Slack: `slack-implementation-advisor`. - Tableau Next: `tableau-next-implementation-advisor`. The product-specific skills are candidate boundaries until specialist depth and live tests establish production readiness. State that limitation when it matters. Invoke platform constraints, packaging, security, and operations as cross-cutting checks; this skill owns the final recommendation. ## Security Review handoff When Security & Readiness owns the interview, provide the implementation's required and optional data flows, identities, permissions, secrets boundary, tenancy, external services, APIs, integrations, clients, and unresolved design risks. For Agentforce, add action audience and public/private classification, action execution identity and authorization evidence, prompt data sources and injection mitigations, and any Connected App/ECA OAuth boundary. Follow the customer and their data: authentication or Salesforce data exchange makes an external endpoint part of the candidate review scope. Product specialists remain unchanged in this release; route their supplied evidence through this contract rather than inventing product-specific review rules. ## Evidence and delegation Prefer practitioner knowledge, then current documentation and approved live evidence. Preserve conflicts for review. Label facts, practitioner knowledge, inferences, recommendations, assumptions, and unknowns. Never reproduce generic Salesforce build, deployment, or CLI skills.
View on GitHub