Skip to content

PranavNagrecha/AwesomeSalesforceSkills / skills/admin/configuration-workbook-authoring

configuration-workbook-authoring

Verified

Author the Salesforce Configuration Workbook — the structured, reviewable handoff document an admin uses to execute a feature across Objects/Fields, Page Layouts, Profiles/PSGs, Sharing, Validation, Automation, List Views, Reports, Integrations, and Data. Triggers: 'salesforce configuration workbook', 'admin handoff document', 'implementation workbook'. NOT for object design itself (use admin/custom-field-creation, admin/lookup-and-relationship-design, agents/object-designer/AGENT.md), NOT for permission set design (use admin/permission-set-architecture, agents/permission-set-architect/AGENT.md), NOT for Flow construction (use skills/flow/* and agents/flow-builder/AGENT.md), and NOT for the deployment manifest (use skills/devops/metadata-api-retrieve-deploy). More triggers: workbook row schema, recommended_agent, recommended_skills, row_id, source_req_id, RTM linkage block, not-in-scope-this-release, descope ledger, version-lock the workbook, check_workbook.py.

references/examples.md

View on GitHub
# Examples — Configuration Workbook Authoring

These are workbook *excerpts* — never the whole 10-section file. Each excerpt
shows the row schema in action so authors can see how `source_req_id`,
`recommended_agent`, and `recommended_skills[]` get populated.

---

## Example 1: Sales Cloud Greenfield — Objects + Fields Section

**Context:** Greenfield Sales Cloud rollout for a manufacturing distributor.
Fit-gap analysis surfaced a need for a custom `Account_Plan__c` object plus
two new fields on Opportunity for ATV (Annual Target Volume) and Renewal
Probability. Three user stories drive these (US-2031 plan tracking, US-2032
pipeline target, US-2033 renewal forecast).

**Problem:** Without a structured workbook, the BA hands the admin a Notion
page with bullet points. The admin builds Account_Plan__c with the wrong
parent relationship, picks an inconsistent suffix on the Opportunity fields,
and skips the External ID required by the integration team.

**Solution (workbook excerpt):**

```markdown
## Section 1 — Objects + Fields

| row_id | target_value | owner | source_req_id | source_story_id | recommended_agent | recommended_skills | status | notes |
|---|---|---|---|---|---|---|---|---|
| CWB-OBJ-001 | New custom object `Account_Plan__c`. Master-detail to Account. Auto Number name `AP-{YYYY}-{0000}`. | A. Singh | FG-014 | US-2031 | object-designer | admin/object-creation-and-design; admin/custom-field-creation; data/external-id-strategy | proposed | Cite `templates/admin/naming-conventions.md` for label set. |
| CWB-OBJ-002 | New field `Account_Plan__c.External_Id__c` Text(40), unique, case-insensitive, External ID. | A. Singh | FG-014 | US-2031 | object-designer | data/external-id-strategy; admin/custom-field-creation | proposed | Required for upsert from CRMA pipeline. |
| CWB-OBJ-003 | New field `Opportunity.Annual_Target_Volume__c` Currency(16,2). | A. Singh | FG-015 | US-2032 | object-designer | admin/custom-field-creation; admin/standard-object-quirks | proposed | Confirm no existing `ATV__c` first via probe. |
| CWB-OBJ-004 | New field `Opportunity.Renewal_Probability__c` Percent(3,0), 0–100. | A. Singh | FG-016 | US-2033 | object-designer | admin/custom-field-creation; admin/formula-fields | proposed | Validation rule lives in CWB-VR-007. |
```

**Why it works:** Every row has both `source_req_id` and `source_story_id`,
exactly one `recommended_agent`, at least one entry in
`recommended_skills[]`, and a non-placeholder `status`. The
`object-designer` agent can pick up rows CWB-OBJ-001 through CWB-OBJ-004 in a
single batch — every cell it needs is on the row.

---

## Example 2: Service Cloud Expansion — Profiles + Permission Sets + PSGs Section

**Context:** Service Cloud expansion adding a Tier-2 Support persona. The
existing org has `Feat_KnowledgeReader` and `Obj_CaseEditor` Permission Sets;
fit-gap calls for adding a `Feat_OmniSupervisor` Feature PS, composing a new
PSG `Tier2Support_Bundle`, and muting `View All Cases` from the bundle for
NA-region T2 reps only.

**Problem:** The previous workbook attempt collapsed PSG composition and
muting into one row, addressed it to "the admin team" with no agent, and
didn't reference the muting decision. The permission-set-architect agent
couldn't pick it up cleanly.

**Solution (workbook excerpt):**

```markdown
## Section 3 — Profiles + Permission Sets + PSGs

| row_id | target_value | owner | source_req_id | source_story_id | recommended_agent | recommended_skills | status | notes |
|---|---|---|---|---|---|---|---|---|
| CWB-PSG-021 | New Feature PS `Feat_OmniSupervisor` granting Omni-Channel supervisor System Permissions per `templates/admin/permission-set-patterns.md`. | M. Vidal | FG-031 | US-2104 | permission-set-architect | admin/permission-set-architecture; security/permission-set-groups-and-muting | proposed | |
| CWB-PSG-022 | New PSG `Tier2Support_Bundle` composing `Obj_CaseEditor`, `Feat_KnowledgeReader`, `Feat_OmniSupervisor`. | M. Vidal | FG-031 | US-2104 | permission-set-architect | admin/permission-set-architecture; security/permission-set-groups-and-muting | proposed | Deployment order per `devops/permission-set-deployment-ordering`. |
| CWB-PSG-023 | New Muting PS `Mute_NARegion_In_Tier2Support_Bundle` muting `ViewAllCases` from the PSG. | M. Vidal | FG-032 | US-2104 | permission-set-architect | security/permission-set-groups-and-muting | proposed | Apply only when `User.RegionCode__c = NA`. |
```

**Why it works:** Each row is one configurable artifact (one PS, one PSG, one
muting PS). The permission-set-architect agent can execute them in deployment
order. None of the rows accidentally do object or sharing work — those would
be split into Objects+Fields and Sharing Settings rows.

---

## Example 3: Multi-Flow Feature — Automation Section

**Context:** A renewal-forecast feature requires two record-triggered Flows on
Opportunity (one before-save normalization, one after-save notification) plus
a scheduled Flow to roll up renewal probability into a custom report rollup
field on Account. Decision tree resolves to Flow over Apex per
`automation-selection.md` Q3.

**Problem:** A single row "Build the renewal automation" hides the three
distinct triggers, gives the flow-builder agent no decision-tree citation, and
makes it impossible to test or roll back independently.

**Solution (workbook excerpt):**

```markdown
## Section 6 — Automation (Flow / Apex / Approvals)

| row_id | target_value | owner | source_req_id | source_story_id | recommended_agent | recommended_skills | status | notes |
|---|---|---|---|---|---|---|---|---|
| CWB-AUT-041 | Record-triggered Flow `Opportunity_Renewal_BeforeSave_v1` on Opportunity, before-save, normalize Renewal_Probability__c into bucket field. | J. Park | FG-051 | US-2033 | flow-builder | flow/record-triggered-flow-patterns; flow/flow-record-save-order-interaction | proposed | Per `automation-selection.md` Q3 → Flow (no callout, no DML beyond same-record). |
| CWB-AUT-042 | Record-triggered Flow `Opportunity_Renewal_AfterSave_Notify_v1` on Opportunity, after-save, send custom notification to Owner when Renewal_Probability__c crosses 60%. | J. Park | FG-052 | US-2033 | flow-builder | flow/record-triggered-flow-patterns; flow/fault-handling | proposed | Fault path required; cite `templates/flow/FaultPath_Template.md`. |
| CWB-AUT-043 | Scheduled Flow `Account_Renewal_Rollup_Daily_v1`, runs nightly, rolls up max(Renewal_Probability__c) of open Opportunities into Account.Best_Renewal_Probability__c. | J. Park | FG-053 | US-2033 | flow-builder | flow/scheduled-flows; flow/flow-bulkification | proposed | Confirm RSF alternative not viable in CWB-OBJ note; rollup field requires Number(3,0). |
```

**Why it works:** Each Flow is its own row. Each row cites the
`automation-selection.md` decision tree branch (or links to the alternative
considered). The flow-builder agent picks up one row at a time, builds one
Flow at a time, and reports back independently per row.

---

## Anti-Pattern: The "Build Sheet" Wiki

**What practitioners do:** Author a free-text Notion or Google Doc titled
"Q2 Build Sheet" with bullet points like "add Account Plan object", "PSGs
for tier 2", "renewal flow stuff". No row IDs, no `source_req_id`, no agent
field.

**What goes wrong:** The doc is untestable. The admin reads it, builds what
they think it means, and the QA team approves what they think *that* means.
Six months later nobody can answer "which fit-gap row drove this field?". The
RTM is fiction.

**Correct approach:** Use the row schema. If a change is too vague to write
as a single row with all canonical fields populated, it is not yet ready for
the workbook — it goes back to the BA as a fit-gap clarification.

Here is the same wiki bullet run through that test. "PSGs for tier 2" becomes
one row per artefact in the `cwb.json` envelope, and the three fields the
bullet had no answer for are the three that force the clarification:

```json
{
  "row_id": "CWB-PSG-022",
  "section": "Profiles + Permission Sets + PSGs",
  "target_value": "New PSG `Tier2Support_Bundle` composing `Obj_CaseEditor`, `Feat_KnowledgeReader`, `Feat_OmniSupervisor`",
  "owner": "M. Vidal",
  "source_req_id": "REQ-031",
  "source_story_id": "US-2104",
  "recommended_agent": "permission-set-architect",
  "recommended_skills": [
    "admin/permission-set-architecture",
    "security/permission-set-groups-and-muting"
  ],
  "status": "proposed",
  "notes": "Deployment order per devops/permission-set-deployment-ordering. Muting PS is CWB-PSG-023, a separate row."
}
```

`owner`, `source_req_id` and `source_story_id` are the ones a wiki bullet
never carries — which is the whole reason the bullet reviews cleanly and the
row does not. `references/worked-examples.md` walks the format rules that
govern the other nine columns.
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall