Skip to content

forcedotcom/sf-skills / skills/nonprofit-volunteer-management-records-create

nonprofit-volunteer-management-records-create

Verified

Use this skill to write Salesforce Nonprofit Cloud Volunteer Management (VM) records to an org once the user knows what to create, from a confirmed plan or a list of records: it reuses existing records by their Ids instead of duplicating them, asks for any missing required values, shows each object's records table for approval, inserts the records in dependency order, reads each write back, recovers from write errors, and reports every created Id. TRIGGER when a user wants to create, insert, import, or write VM records. DO NOT TRIGGER to decide what a program needs (use nonprofit-volunteer-management-structure-generate), to explain what a VM object or field is for (use nonprofit-volunteer-management-model-get), to enable Volunteer Management in an org, for per-person records (JobPositionAssignment, PersonLocationAvailability, PersonCompetency / PersonExamination, ConstituentRole), or for NPSP / Volunteers for Salesforce (V4S), which is unsupported.

FileKindSize
SKILL.mdskill md15.7 KB
references/field-guide.mdmarkdown24.7 KB
references/mcp-invocation.mdmarkdown14.2 KB
references/review-format.mdmarkdown8.3 KB
scripts/expand-series.pyexecutablescript28.7 KB
scripts/render-review-table.pyexecutablescript17.5 KB
scripts/tests/test_expand_series.pyexecutablescript27.8 KB
scripts/tests/test_render_review_table.pyscript20.4 KB

references/field-guide.md

View on GitHub
# Volunteer Management Record Creation — Field Guide

The field reference for creating Volunteer Management records: which fields exist, what each
means when creating a record, which to set (the tiers below), and their defaults. This file covers
how to fill fields; what an object means and how the model connects is in
`nonprofit-volunteer-management-model-get`.

## Where field facts come from

- **Standard fields** (API names, required flags, tiers, meaning) come from **this file**. Never
  describe an object to learn its standard fields, and don't make "checking the standard fields" a
  visible step.
- **A description is used for three things:** required **custom** fields the org added, **live
  picklist values**, and grounding a fix for a write error. When a description is in hand and
  differs from this file (an admin-renamed label, a picklist value, a required flag), the
  description wins; API names don't change on a rename, labels do.
- **If the user declines the read**, this file alone still covers every standard field. Required
  custom fields and bad picklist values then surface as write errors and are fixed in SKILL.md
  step 9.
- **Query only fields this file documents.** A column that isn't in an object's table doesn't
  exist for this skill — e.g. Operating Hours has no `Type` field, and selecting one fails the
  whole query with `INVALID_FIELD`.
- **Every reuse query is bounded** to the planned records (`Name IN (…)`, or `Name LIKE` a
  distinctive stem when the org may word it differently, e.g. `Name LIKE '%Forklift%'`). Never
  read a whole object. Reuse keys:
  - Volunteer Initiative, Location, Competency, Examination, Position: `Name`.
  - Job Position: `Name` + `VolunteerInitiativeId`.
  - Operating Hours: `TimeZone` + the set of Time Slot windows (`DayOfWeek` + `StartTime` +
    `EndTime`). Never the name.
  - Job Position Shift: `JobPositionId + StartDate + EndDate + StartTime + EndTime`.

  The calls, and the apostrophe rule, are in `references/mcp-invocation.md`.

## Tiers

The tier codes are internal; the user sees "required", "recommended" or "optional".

| Tier | Meaning | Action |
|------|---------|--------|
| **T1** | Always set | Every record needs it (required, or an identifier/relationship without which the record is invalid or useless). |
| **T2** | Set by default; explain if skipping | Set it unless the user opts out, and say what skipping costs. |
| **T3** | Set only if the program calls for it | Leave blank unless the plan or the user needs it. |
| **T4** | Never set — system-managed | Filled by the platform, a formula, an auto-number or a Data Processing Engine; writing it errors or is overwritten. |
| **T5** | Never mention | Belongs to another feature; irrelevant to creation. |

**Fill what's known; ask before inventing.** Fill values the plan or the user already gave. For a
field with no established value, propose one and ask; never invent a value the user didn't give.
Status may be proposed as `Upcoming`. **Location Type has no default and is always asked** (or
copied from the paired Location on an Address).

**`Description` (every object) — summarize from context.** Summarize what you already know about
that record (the plan on a chained run, or the conversation). When there's nothing record-specific,
leave it blank. That's why Volunteer Initiative and Position carry `Description` at T2 (a headline
record almost always has context) and every other object at T3.

**`Req` vs `Tier` differ on purpose.** `Req` is what the platform rejects when blank; `Tier` is what
this skill fills anyway. A shift with no `StartDate` is useless, so it is T1 though `Req: No`.

## Picklists

Read every picklist's values from the object description at run time; never assume a set, because
admins change them. Offer every active value.

- **Platform picklists** (Time Zone, Time Slot Day of Week and Type): the values are fixed. A value
  outside them is rejected before the write, never sent.
- **Every other picklist:** if no live value fits, ask whether to add one. Adding a value needs its
  own yes from the user (`references/mcp-invocation.md` → Adding a picklist value): on headless-360
  the user adds it in Setup, on the Salesforce CLI you add it for them. A value that isn't in the
  list yet is rejected on write, so add it first.

---

## Creation scope

This skill creates the **program structure**. The tables below cover every object it writes.
Per-person and operational records (assignments, availability, a person's qualifications, rollups)
are out of scope — see [Out of scope](#out-of-scope--never-created).

### VolunteerInitiative — the program container (top of the graph)

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The name of the program — the top-level initiative its job positions, shifts, and assignments roll up under. | — |
| `Status` | Unrestricted picklist | No | **T2** | The lifecycle status of the program. | Propose `Upcoming` for a new program. |
| `Description` | Text(255) | No | **T2** | A description of the program. | Set by default; summarize from context. |
| `StartDate` | Date | No | T3 | The date the program's volunteering starts. | Optional; set together with `EndDate` for a bounded program. |
| `EndDate` | Date | No | T3 | The date the program's volunteering ends. | Optional; set alongside `StartDate` for a bounded program. |
| `IsPublished` | Boolean | No | T3 | Whether the program is visible to volunteers in an Experience Cloud site (default `false`). | Nothing out of the box publishes it — set only if publishing now. |
| `ParentVolunteerInitiative` (`ParentVolunteerInitiativeId`) | Lookup → VolunteerInitiative | No | T3 | The parent program in the initiative hierarchy — engagement rollups aggregate up to parents (max 5 levels, max 1000 child relations). | Write the parent first, then the children in a second call with the parent's **literal Id** — a `@refId` to another Volunteer Initiative in the same call fails with `MALFORMED_ID`. |
| `TotalVolunteerHours`, `TotalActiveVolunteerCount`, `AttendanceRate`, `OpenAssignmentCount`, `FilledAssignmentCount`, `TotalAssignmentCount` | Number | No | **T4** | Data Processing Engine engagement rollups — hours volunteered, active volunteers, attendance rate, and open/filled/total assignment counts. | Empty containers until you build the automation; writing them is meaningless/overwritten — never set. |

### Location — the physical/virtual site

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The name of the site where the volunteer work happens — what job positions link to and matching compares against. | — |
| `LocationType` | Unrestricted picklist | Yes | **T1** | The kind of site. | **No default — always ask.** Offer the live values; if none fits, offer to add one. |
| `TimeZone` | Platform picklist (IANA) | No | **T2** | The site's time zone — it feeds the shift and Operating-Hours time zones used in time-based matching. | Set by default when shifts/matching are time-based — it feeds the shift and Operating-Hours time zones; skipping it breaks time-correct scheduling. |
| `VisitorAddress` (`VisitorAddressId`) | Lookup → Address | No | **T2** (new site) | A lookup to the site's postal `Address` record. | A Location has **no inline postal fields**, so the street address lives in a paired `Address`: create the **Location first**, then the Address, then back-fill this lookup. See the **Address** entity below for the fill policy and the 3-step circular write. |
| `Description` | Text | No | T3 | A brief description of the site. | Summarize from context; leave blank if none. |

Query for an existing Location by name and reuse it before creating one. A virtual Location has no Address.

### Address — the paired postal address for a Location

Every **new onsite** Location gets a paired `Address`; that's where the street address lives
(`Location` itself has none). A virtual Location gets none unless the user asks. A **reused**
Location: read its `VisitorAddressId` — if it's set, the Location already has an Address, so don't
create one; if it's empty, offer one.

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Parent` (`ParentId`) | Master-detail → Location | Yes | **T1** | The site (Location) this address is the postal address for. | Set to the paired Location's **literal Id**. Master-detail, so it can't be moved to another Location later. See the 3-step write below. |
| `LocationType` | Unrestricted picklist | Yes | **T1** | The kind of site — mirrors the paired Location's type. | Required, **no default** — **copy the paired Location's `LocationType`**; it must come from the Location, never assumed. |
| `Street`, `City`, `State`, `PostalCode`, `Country` | Text | No | **T1** (user-supplied) | The site's postal address — the reason this record exists (a Location has no inline street fields). | **Ask the user; never fabricate.** Populate whatever they give; a sub-field they omit stays blank. |
| `AddressType` | Unrestricted picklist | No | **T2** | The kind of address. | Offer the live values; if none fits, offer to add one or leave it blank. |
| `TimeZone` | Platform picklist (IANA) | No | **T2** | The site address's time zone. | **Copy the Location's `TimeZone`** — matters for organizations spanning zones. |
| `Description` | Text | No | T3 | A brief description of the site's address. | Summarize from context; blank if none. |
| `Latitude`, `Longitude` | Double | No | T3 | The geographic coordinates of the site's address. | Copy from the Location **only if** it carries geo values; else skip. |
| `DrivingDirections` | Text | No | T3 | Directions to the site. | Copy from the Location if present. |
| `GeocodeAccuracy` | Picklist | No | **T4** | The accuracy level of the site address's geocode. | Geocoding-managed system value — never hand-set. |

**3-step write.** `Address.ParentId` → Location is required and `Location.VisitorAddressId` →
Address points back, so one call can't create the pair. **(1)** Write the **Location** (no
`VisitorAddressId`); **(2)** write the **Address** with `ParentId` = the Location's literal Id;
**(3)** update the Location's `VisitorAddressId` to the Address's Id. Calls in
`references/mcp-invocation.md`.

### OperatingHours — an availability schedule (matching)

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The name of the availability schedule — the set of weekly windows (its TimeSlots) a volunteer's availability is matched against. | Time zone + a short window-set descriptor — see *Naming* below. |
| `TimeZone` | Platform picklist (IANA) | Yes | **T1** | The time zone the schedule's TimeSlots are read in; matching uses it to align a volunteer's availability with shift times. | Required and **not safely defaultable** — ask (or take from the Location's time zone). |
| `Description` | Text | No | T3 | A description of the schedule. | Summarize from context; blank if none. |

**Create Operating Hours and Time Slots only when the user explicitly asks.** Nothing links them
to a Location or Job Position; a volunteer's own availability record (out of scope) points at them
at runtime.

**Reuse.** One Operating Hours can serve many volunteers. Before creating one, query for an
existing Operating Hours with the same `TimeZone` and the same set of Time Slot windows, and reuse
it. The name never decides reuse.

**Naming.** Time zone + a short window-set descriptor (`America/Los_Angeles — Mornings`,
`America/New_York — Weekends`), so schedules that share a time zone stay distinct. A site label
(`Raceway Hours`) is fine as a human-readable tag, but the name never binds the record to a site.
Use a bare time-zone name only when that zone has a single schedule.

### TimeSlot — a weekly availability window (child of OperatingHours)

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `OperatingHours` (`OperatingHoursId`) | Master-Detail → OperatingHours | Yes | **T1** | The availability schedule this window belongs to. | Set to the parent's **literal Id**. **Write the Operating Hours first, then its Time Slots in a second call.** |
| `DayOfWeek` | Platform picklist | No | **T1** | The day of the week the volunteer is available (`Sunday`…`Saturday`). | Defaults to `Sunday` when unset — rarely the intended day, so **ask**. |
| `StartTime` | Time | Yes | **T1** | The time the availability window starts. | No default — ask. |
| `EndTime` | Time | Yes | **T1** | The time the availability window ends. | Must be ≥ `StartTime`. No default — ask. |
| `Type` | Platform picklist | No | T3 | The kind of time slot (`Normal` or `Extended`). | Defaults to `Normal`; Volunteer Management doesn't use it — skip unless the user asks for `Extended`. |
| `TimeSlotNumber` | Auto-number | System | **T4** | The record's auto-generated name (autonumber mask `TS-{0000}` — e.g. `TS-0001`). | Platform-assigned (`createable=false`); never set. |

**No duplicate windows.** Within an OperatingHours, don't create a TimeSlot that duplicates an
existing window (same `DayOfWeek` + `StartTime` + `EndTime`). Reuse the parent OperatingHours and
its existing TimeSlots instead of re-creating an identical set.

### Position — reusable role template (optional)

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The name of the reusable role template that postings are created from. | — |
| `Status` | Unrestricted picklist | No | **T2** | The status of the role template. | Propose `Upcoming`. |
| `Description` | Text(255) | No | **T2** | A description of the role. | Set by default. |
| `IsInternalApprovalRequired` | Boolean | No | T3 | Whether volunteer signup for postings that use this role requires internal approval (default `false`). | Does nothing until the customer sets up an Approval Process. |
| `Type` | Unrestricted picklist | No | T3 | The type of role. | Skip unless the org has values or the user asks to add one. |
| `Code` | Text | No | T3 | A short code for the role. | — |

Create a Position when a role is shared across postings or programs, including the same role posted at several Locations. Put a qualification every posting of that role needs on the Position once, not on each Job Position.

### PositionQualification — a requirement on the template

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Position` (`PositionId`) | Master-Detail → Position | Yes | **T1** | The role template this requirement belongs to; matching checks it for every Job Position that uses the Position. | Set to the Position's literal Id — write the Position first. |
| `QualificationReferenceRecord` (`QualificationReferenceRecordId`) | Lookup {Competency, Examination} | Yes | **T1** | The Competency or Examination a volunteer must have to fill the role. | Matching uses only Competency requirements; an Examination requirement is informational. |
| `QualificationLevel` | Integer | No | **T2** | The qualification level the role requires — matching compares a volunteer's level against it. | Set it whenever matching is wanted: a Competency requirement without a level is skipped by matching. |
| `ProficiencyLevel` | Unrestricted picklist | No | T3 | A descriptive proficiency label for the requirement. | Not used in matching. |

### JobPosition — a concrete posting (role × site)

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The name of the posting volunteers browse and sign up for. | — |
| `Title` | Text(255) | Yes | **T1** | The title of the posting. | Also required — **both `Name` and `Title` must be set.** Can hold the same value or differ (e.g. an internal-facing name vs. an external/public one). |
| `Status` | Unrestricted picklist | No | **T2** | The lifecycle status of the posting. | Propose `Upcoming`. |
| `Location` (`LocationId`) | Lookup → Location | No | **T2** | The site where this posting's work happens. | Optional; set it when the work has a site — it drives time and place matching. |
| `VolunteerInitiative` (`VolunteerInitiativeId`) | Lookup → VolunteerInitiative | No | **T2** | The parent program this posting rolls up to. | **API name is `VolunteerInitiativeId`** (label "Related Volunteer Initiative") — *not* `RelatedVolunteerInitiative` (that field is on JobPositionAssignment). |
| `Position` (`PositionId`) | Lookup → Position | No | T3 | The reusable role template this posting is created from. | Only when a Position template is in use. Its qualifications are checked by matching; nothing is copied. |
| `Description` | Text Area(255) | No | T3 | A description of the posting shown to volunteers. | Summarize from context; blank if none. |
| `StartDate` / `EndDate` | Date | No | T3 | The posting's active window. | — |
| `Manager` (`ManagerId`) | Lookup → Account | No | T3 | The posting's manager (an Account). | — |
| `Code` | Text(255) | No | T3 | A short code for the posting. | — |
| `LastFilledDate` | Date | No | T3 | The date the posting was last filled. | Nothing fills it automatically. A new posting hasn't been filled, so leave it blank unless the user gives a date. |

### JobPositionQualification — a posting-specific requirement

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `JobPosition` (`JobPositionId`) | Master-Detail → JobPosition | Yes | **T1** | The posting this requirement applies to. | Set to the Job Position's literal Id — write the Job Position first. |
| `QualificationReferenceRecord` (`QualificationReferenceRecordId`) | Lookup {Competency, Examination} | Yes | **T1** | The Competency or Examination this posting requires, in addition to its Position's requirements. | Matching uses only Competency requirements; an Examination requirement is informational. Ask. |
| `QualificationLevel` | Integer | No | **T2** | The qualification level this posting requires — matching compares a volunteer's level against it. | Set it whenever matching is wanted: a Competency requirement without a level is skipped by matching. |
| `ProficiencyLevel` | Unrestricted picklist | No | T3 | A descriptive proficiency label for the requirement. | Not used in matching. |
| `Description` | Text(255) | No | T3 | A description of the requirement. | From context; blank if none. |

Matching uses QualificationLevel on Competency qualifications. A Job Position's own qualifications and its Position's qualifications are checked together, keeping the higher level per Competency. Nothing is copied from the Position to the Job Position — don't add a Job Position Qualification that repeats one on its Position.

### JobPositionShift — a dated, time-bounded slot

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `JobPosition` (`JobPositionId`) | Lookup → JobPosition | Yes | **T1** | The posting this shift is a dated, time-bounded slot of. | Set to the Job Position's literal Id. |
| `StartDate` | Date | No | **T1** | The date the shift takes place. | **No `StartDateTime` field exists** — date and time are separate. Ask. |
| `EndDate` | Date | No | **T1** | The shift's end date. | Equal to `StartDate` on a same-day shift; **later when the shift crosses midnight or spans days** (StartDate/EndDate are independent DATE fields). |
| `StartTime` | Time | No | **T1** | The time the shift starts. | — |
| `EndTime` | Time | No | **T1** | The time the shift ends. | Must be after `StartTime` **only on a same-day shift** (`EndDate == StartDate`); when the shift ends on a later `EndDate` the end time may be earlier (e.g. 23:00 → 03:00 next day). |
| `TimeZone` | Platform picklist (IANA) | No | **T1** | The shift's time zone. | No safe default — take it from the Location, or ask. |
| `Status` | Unrestricted picklist | No | **T2** | The lifecycle status of the shift. | Propose `Upcoming`; `scripts/expand-series.py` writes `Upcoming` when the spec has no `status`. |
| `MaximumAttendeesCount` | Integer | No | T3 | The maximum number of volunteers for the shift. | Set when the shift has a headcount limit. Capacity is counted, never enforced — blocking a full shift needs a customer validation rule. |
| `Description` | Text(255) | No | T3 | A description of the shift. | Summarize from context; blank if none. |
| `Name` | Auto-number | System | **T4** | The record's auto-generated name (autonumber mask `JPS-{0000000000}` — e.g. `JPS-0000000001`). | Platform-assigned (`createable=false`); never set. |
| `PositionAssignmentCount` | Integer | No | **T4** | The number of volunteers assigned to the shift. | Maintained by the platform as assignments are added — never set. |
| `RemainingCapacity` | Formula | No | **T4** | The remaining open spots on the shift. | Formula: when both counts are set and the maximum is above 0, `MAX(0, MaximumAttendeesCount − PositionAssignmentCount)`; otherwise blank. Never set. |
| `RecurrenceSchedule` (`RecurrenceScheduleId`) | Lookup → RecurrenceSchedule | No | **T5** | A lookup to the recurrence schedule the shift belongs to. | Owned by Volunteer Management's own recurrence feature — never create or link; write one shift per occurrence instead. |

Bound every series with an end date or a count (`count` in `scripts/expand-series.py`); never an open-ended set.

### Competency — a skill a qualification can require

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The competency name — what qualifications and matching point at. | Ask. |
| `Category` | Unrestricted picklist | No | T3 | The category of the competency. | Skip unless the org has values or the user asks to add one. |
| `Description` | Text Area(1000) | No | T3 | A description of the competency. | From context; blank if none. |

Query for an existing Competency by name and reuse it before creating one.

### Examination — a credential a qualification can require

| Field | Type | Req | Tier | Meaning | Policy |
|-------|------|-----|------|---------|--------|
| `Name` | Text (Name) | Yes | **T1** | The examination or credential name. | Ask. |
| `Status` | Unrestricted picklist | No | T3 | The status of the examination. | Optional; offer the live values. |
| `Description` | Text Area(255) | No | T3 | A description of what the credential verifies. | From context; blank if none. |

Examinations are not used in matching; query for an existing one by name and reuse it before creating one.

---

## Out of scope / never created

Created at runtime or during volunteer onboarding, or part of the person data model Volunteer
Management builds on. This skill never writes them; if the user asks, say so.

| Entity | Why it's not created here |
|--------|---------------------------|
| `RecurrenceSchedule` | Volunteer Management's own recurrence feature owns it; write discrete `JobPositionShift` records instead. |
| `JobPositionAssignment` | A person doing a shift — operational, created when a volunteer signs up. |
| `PersonLocationAvailability` | A volunteer's own availability (ties a Location + OperatingHours) — set during onboarding. |
| `PersonCompetency` / `PersonExamination` | A person's own qualifications — recorded per volunteer. |
| `ConstituentRole` | Marks a Person Account as a volunteer — per-person onboarding. |
| `DonorGiftSummary` | A Data Processing Engine rollup target, not authored directly. |
| `Account` / `Person Account` | The volunteer themselves — a person record created during onboarding, not by program-structure setup. |
| `PartyRelationshipGroup` | Designates an Account as a household — person data model. |
| `AccountAccountRelation` / `AccountContactRelation` / `ContactContactRelation` | Relationship links between people and organizations — person data model. |

---

## Creation limits & validation rules

Enforce/respect these when generating the graph:

- **Volunteer Initiative hierarchy:** max 5 levels of `ParentVolunteerInitiativeId`; max 1000 child relations.
- **Shift generation:** a tree call takes at most 200 records. Expand every series with `scripts/expand-series.py`; over 200 records it needs `--out-dir`, which writes one body per 200-record batch.
- **JobPositionShift:** end datetime ≥ start datetime — i.e. `EndDate ≥ StartDate`, with `EndTime > StartTime` required **only when `EndDate == StartDate`** (a shift that crosses midnight or spans days has a later `EndDate`, so its `EndTime` may be earlier); `MaximumAttendeesCount ≥ 0`.
- **JobPositionAssignment:** end datetime ≥ start datetime.
- **Capacity:** counted, never enforced — nothing out of the box blocks a signup when a shift is full. A customer validation rule may reject writes; handle its error in SKILL.md step 9.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall