@springbrand/cli 0.1.0

package/skill/springbrand-gtm/SKILL.md

@@ -0,0 +1,34 @@
+---
+name: springbrand-gtm
+description: Use SpringBrand for substantive go-to-market and growth work such as market research, positioning, campaigns, creator or partner outreach, distribution, and conversion analysis.
+---
+
+# SpringBrand GTM
+
+Use this Skill when the user asks for substantive marketing or growth work,
+even when they do not name SpringBrand. It is a scenario guide, not a separate
+MCP executor. Use the common SpringBrand Skill to discover current operations
+and then hand off to the Domain Skill that owns the actual effect.
+
+## Recognize the work
+
+Typical requests include competitor or audience research, demand signals,
+positioning, SEO or paid acquisition, creator and partner discovery, outreach,
+campaign assets, distribution, conversion, and performance analysis. Simple
+proofreading or unrelated coding is not GTM work.
+
+## Workflow
+
+1. Restate the business goal, audience, constraints, and desired outcome.
+2. Use `search_tools` through the common MCP guidance to discover current
+   Plugins or Action APIs. Do not rebuild a capability that the Gateway can
+   provide.
+3. Choose one current workflow owner: Platform for Plugin and Creation work,
+   Action API for a SpringBrand-managed service, or Connector for a connected
+   third-party source.
+4. Follow that Domain Skill's contract, risk, approval, and execution rules.
+5. Report evidence, limitations, and next actions without claiming an
+   operation ran when it was only discovered.
+
+GTM discovery does not grant permission or authorize execution. Unknown writes
+remain `outcome_unknown` and are never retried automatically.

package/skill/springbrand-platform/SKILL.md

@@ -0,0 +1,55 @@
+---
+name: springbrand-platform
+description: Use SpringBrand Platform for Plugin discovery and lifecycle work, Artifact creation, uploads, and Creation publication. Use this guidance after the common SpringBrand MCP Skill identifies a Platform operation.
+---
+
+# SpringBrand Platform
+
+Use the unified SpringBrand MCP for Plugin, Artifact, and Creation workflows.
+The CLI has already configured and authenticated the MCP connection; this
+Skill does not install a Plugin, configure an endpoint, or start authentication.
+
+## Operation rules
+
+Use `search_tools` once with a concise English query that preserves the
+requested Plugin, Artifact, Creation, format, and publication goal. Select a
+Platform operation from the returned results and copy its opaque Tool ID
+unchanged. Use `get_tool_schemas` for the current contract before executing.
+Ignore results whose effect belongs to Action API or Connector and hand the
+user back to the common Skill for an explicit workflow transition.
+
+`get_tool_schemas` and discovery do not add a Plugin, upload an Artifact, or
+publish a Creation. Treat the current contract as authoritative and ask for
+missing business values instead of inventing them.
+
+## Plugin lifecycle
+
+1. Discover the requested Plugin operation and inspect its current contract.
+2. Present the current Plugin identity, version, usage instructions, access
+   state, and acquisition requirement when available.
+3. For a paid or uncertain acquisition, explain the cost and wait for the
+   user's explicit agreement. Never purchase or bypass host approval.
+4. Retrieve distribution content when the user needs instructions or
+   components. Retrieval supplies content; it does not install or execute the
+   Plugin.
+5. Remove or rate only when the user explicitly asks.
+
+## Artifact and Creation workflow
+
+Keep local source files, credentials, caches, and State Documents out of an
+Artifact. Before upload, verify the intended format, referenced assets, file
+names, MIME types, sizes, and total bytes. Use the exact upload contract and a
+stable idempotency key.
+
+Uploading and publishing are separate writes. Explain the effect and any
+reported risk before each confirmation gate. Execute once with
+`execute_tools`, then use `get_execution` only with an exact returned
+execution ID. An unknown write result is not failure and is never retried
+automatically; preserve its idempotency identity and verify the effect.
+
+## Boundaries
+
+Platform owns Plugin lifecycle, Artifact, Creation, upload, and publication.
+It does not manage third-party account connections or execute an Action API.
+Use one executor at a time and preserve task state when handing off to another
+Domain Skill.

package/skill/springbrand-platform/references/plugin-discovery.md

@@ -0,0 +1,94 @@
+# Platform and Plugin discovery through the unified Meta Tools
+
+Read this reference when the Platform Skill needs to find a Plugin or Platform
+operation. It records Agent behavior only; the runtime MCP schema and returned
+current contract are authoritative.
+
+## One bounded search
+
+Use `search_tools` with one concise English query. Translate a non-English
+request yourself while preserving explicit Artifact type, subject, workflow,
+audience, and desired output constraints. Remove SpringBrand host names,
+courtesy language, and generic words that do not distinguish the needed Plugin
+or Platform operation.
+
+Issue one search, not a synonym fan-out. The response is one bounded list. It
+is not globally ranked, not a complete Marketplace, and has no search
+pagination for the Agent to invent.
+
+Useful query shapes include:
+
+- `[artifact type] [subject] [workflow]`
+- `retrieve SpringBrand Plugin for [specific task]`
+- `upload SpringBrand Creation`
+- `publish SpringBrand Creation version`
+- `list my SpringBrand Creations`
+
+Use only constraints the user actually supplied. Do not inject a category,
+format, audience, or publication intent merely to improve recall.
+
+## Interpret results
+
+Preserve returned order while checking actual fit. A result is compatible only
+when its description and requirements can satisfy the user's explicit task.
+Scores or list position are not task-success probabilities.
+
+Keep these outcomes distinct:
+
+- **Fitting result** — preserve its opaque Tool ID exactly and inspect the
+  current contract with `get_tool_schemas`.
+- **Complete empty or incompatible result** — report no fit for this query, or
+  revise it once when a more precise version follows directly from the user's
+  stated constraints.
+- **Incomplete result** — it cannot establish no-match. Narrow or revise the
+  query; do not claim that the Marketplace has no fit.
+- **Provider or tool failure** — report the actual transport, authentication,
+  permission, schema, or upstream failure. An error is not a no-match and must
+  not trigger a fallback presented as success.
+
+Unified discovery can include Action API and Connector operations. Ignore them
+inside Platform. If the user's real task belongs there, use an explicit Domain
+Transition rather than executing across domains.
+
+## Plugin retrieval is not Plugin use
+
+A discovered Plugin operation may retrieve instructions, components, or files.
+That effect does not:
+
+- install or run the Plugin;
+- add or purchase it;
+- execute a Generated Business Skill or Action;
+- upload or publish the user's Artifact;
+- complete the task described by the retrieved content.
+
+Follow retrieved usage instructions only within the user's authorized task and
+the Platform Skill's cost, risk, and Domain boundaries.
+
+## Current contract and related operations
+
+Send exact discovered Tool IDs to `get_tool_schemas`. The current contract is
+the authority for description, input/output schemas, risk, known cost, access
+state, and related operations.
+
+The Plugin lifecycle may need distinct effects: inspect current Plugin facts,
+add, retrieve a distribution, remove, or rate. Upload, list Creations, and
+publish a selected version are also distinct Platform effects. Use the exact
+opaque Tool ID returned for each effect; never derive one ID from another,
+from a Plugin title, or from this reference.
+
+Contract inspection does not execute or authorize an operation. A failed
+lookup does not prove the Plugin or operation is absent and does not authorize
+silently switching to another candidate.
+
+## Execute within the lifecycle
+
+Call `execute_tools` only after the Platform Skill has applied the selected
+operation's current schema, access, cost, risk, and confirmation rules. Every
+logical invocation uses one stable idempotency key retained before the call.
+The current operation description defines success: retrieving content is not
+equivalent to applying it, adding is not purchasing, upload is private, and
+publication is a separate public action.
+
+Only use `get_execution` when execution returns an exact non-null execution
+ID. A synchronous Platform result cannot be polled. Reading a saved result does
+not repeat the operation or authorize a follow-up action.

package/skill/use-springbrand/SKILL.md

@@ -0,0 +1,70 @@
+---
+name: use-springbrand
+description: Use SpringBrand through the configured MCP server when the user needs a Platform operation, an Action API, or a connected third-party service. Discover the operation with search_tools, then use its current schema and the appropriate Domain Skill.
+---
+
+# SpringBrand MCP
+
+Use the configured MCP server named `springbrand` for SpringBrand work. The
+CLI owns installation, endpoint configuration, and API-key authentication. Do
+not ask the user to copy credentials into a prompt or configure a second MCP
+server.
+
+## Start with discovery
+
+When the user asks what SpringBrand can do or asks you to perform an external
+operation, call `search_tools` with one concise English query. Preserve the
+requested service, operation, and important constraints. `search_tools` is a
+bounded relevant-operation search, not a complete catalog; only a complete
+empty result establishes no match.
+
+Use the returned opaque Tool ID unchanged. Call `get_tool_schemas` before an
+execution unless a current contract is already available. Build arguments
+strictly from that schema and ask the user for missing business values.
+
+## The five Meta Tools
+
+| Tool | Use |
+| --- | --- |
+| `search_tools` | Discover relevant operations. |
+| `get_tool_schemas` | Read the current input and output contract. |
+| `manage_connections` | List, start, inspect, repair, or disconnect a third-party account when the Connector workflow requires it. |
+| `execute_tools` | Run one discovered operation with schema-valid arguments and a stable idempotency key. |
+| `get_execution` | Read an exact saved execution result or status. |
+
+The tools are independently composable. Do not invent a fixed sequence when
+the user already has a current Tool ID, schema, connection pointer, or
+execution ID.
+
+## Choose the workflow guidance
+
+- Plugin, Artifact, Creation, upload, or publication work follows
+  `springbrand-platform`.
+- A SpringBrand-hosted API service follows `springbrand-action-api`.
+- Gmail, GitHub, Slack, or another connected account follows
+  `springbrand-connector`.
+- A substantive marketing or growth request may use `springbrand-gtm` to
+  choose the relevant business workflow.
+
+Use one Domain Skill for the current operation. If a task crosses domains,
+finish or hand off the current state explicitly before starting the next
+workflow.
+
+## Execution safety
+
+- Copy Tool IDs, execution IDs, connection IDs, and authorization attempt IDs
+  exactly as returned. Never parse or reconstruct them.
+- Execute only with the current contract, schema-valid arguments, and a stable
+  idempotency key for the logical run.
+- Explain known cost, risk, approval, and access requirements before a write.
+- Treat `outcome_unknown` as uncertain. Preserve its idempotency identity and
+  verify the external effect; never retry automatically.
+- A lookup failure is not proof that an operation failed, and a missing
+  connection is not permission to invent credentials.
+
+## Results
+
+Only `succeeded` means an operation completed. A running or queued execution
+remains in progress; use `get_execution` only with the exact returned ID. Keep
+saved image previews and links as returned by the Gateway and do not fetch an
+original file merely to improve presentation.