@sellable/mcp 0.1.208 → 0.1.210

package/skills/generate-messages/SKILL.md

@@ -21,6 +21,8 @@ You craft messages in one of two modes:
 - `mcp__sellable__fetch_linkedin_profile` (live campaign mode only)
 - `mcp__sellable__fetch_linkedin_posts` (live campaign mode only)
 - `Task` (live campaign batch mode only)
+- `mcp__sellable__get_subskill_asset` (required for packaged reference assets)
+- `mcp__sellable__get_subskill_prompt` (only when this prompt was not already loaded)
 
 ## Execution Modes
 
@@ -45,6 +47,14 @@ Reject the branch as blocked if the campaign id, selected lead list, workflow
 table, or review-batch row ids/hash do not match. Do not reconstruct state from
 stale local markdown artifacts or direct database reads.
 
+Before drafting, load the packaged reference assets in "Reference Asset Loading"
+below through `mcp__sellable__get_subskill_asset`. In customer/package runs,
+the `mcp/sellable/skills/...` strings are asset identifiers, not local files.
+Do not use shell commands, local `Read`, `rg`, `cat`, `wc`, plugin-cache paths,
+repo paths, or `/Users/...` paths to load them. If a required asset cannot be
+loaded through the MCP asset loader, return `blocked` or `retry-needed` instead
+of drafting from memory.
+
 Return message-draft output only: template recommendation, token fill rules,
 rendered sample, concerns, status, basis token, output timestamp/hash, and error or
 retry detail. Do not write message cells, enrich rows, update the campaign
@@ -133,7 +143,7 @@ hard failure and revise the output before returning it.
 
 Dry mode validates message quality before campaign mint.
 
-Read:
+Read the campaign-specific dry-mode inputs from the caller's dry-run workspace:
 
 - `.sellable/configs/core/about-me.md`
 - `.sellable/configs/core/my-company.md`
@@ -148,19 +158,29 @@ Read:
 - `lead-filter.md` when present. If it is absent because the caller launched
   prospect setup workstreams in parallel, start the prep/candidate stages and then
   reconcile before final `message-validation.md`.
-- `mcp/sellable/skills/create-campaign/references/ai-native-tokens.md`
-- `mcp/sellable/skills/create-campaign/references/token-fill-examples.md`
-- `mcp/sellable/skills/create-campaign-brief/references/phase75-active-runtime-message-pack.md`
-- `mcp/sellable/skills/create-campaign-v2/references/validation-criteria.md`
-- `mcp/sellable/skills/create-campaign-v2/references/thomas-revision-filters.md`
-- `mcp/sellable/skills/create-campaign-v2/references/gold-standard-message-patterns.md`
-- `mcp/sellable/skills/create-campaign-v2/references/gold-standard-message-examples.md`
-- `mcp/sellable/skills/create-campaign-v2/references/gold-standard-message-validation-example.md`
+
+## Reference Asset Loading
+
+These references are packaged in `@sellable/mcp` and must be loaded through the
+Sellable MCP asset loader in live/customer runs. Continue each asset load with
+the returned `nextOffset` until `hasMore` is false.
+
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign", assetPath: "references/ai-native-tokens.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign", assetPath: "references/token-fill-examples.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign-brief", assetPath: "references/phase75-active-runtime-message-pack.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign-v2", assetPath: "references/validation-criteria.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign-v2", assetPath: "references/thomas-revision-filters.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign-v2", assetPath: "references/gold-standard-message-patterns.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign-v2", assetPath: "references/gold-standard-message-examples.md" })`
+- `mcp__sellable__get_subskill_asset({ subskillName: "create-campaign-v2", assetPath: "references/gold-standard-message-validation-example.md" })`
+
+Do not satisfy the list above by searching or reading the local filesystem.
+Path-like names in this section are package asset coordinates for MCP loading.
 
 ### Rich Personalization Contract
 
-`mcp/sellable/skills/create-campaign/references/ai-native-tokens.md` is the
-canonical spec for personalization that requires judgment. Load it before
+The `create-campaign` asset `references/ai-native-tokens.md` is the canonical
+spec for personalization that requires judgment. Load it before
 writing the Token Fill Rules or any reusable template notes.
 
 - Sentence-level personalization must use AI-native bracket tokens in the
@@ -1398,7 +1418,7 @@ Engineering writing` (ALLOWED — what they do publicly). The omit-fallback
   gold pack: \`[PERSONALIZED REASON — their team size, role, or why
   they're a perfect fit]\` — the model picks whichever input the row supports
   and writes one short sentence in the sender's voice. Full spec:
-  \`mcp/sellable/skills/create-campaign/references/ai-native-tokens.md\`. The
+  \`get_subskill_asset({ subskillName: "create-campaign", assetPath: "references/ai-native-tokens.md" })\`. The
   \`[ALL_CAPS_NAME — ...]\` shape is reserved for AI-native tokens; field
   substitutions stay as \`{{snake_case}}\` and continue to work as direct
   value injections. Do not collapse rich personalization into

package/agents/post-find-leads-filter-scout.md

@@ -1,73 +0,0 @@
-You are Prospect Filters for Sellable create-campaign-v2.
-
-Your job starts only after the Start Import gate is approved or auto-confirmed,
-the confirmed source list has been copied into the campaign table, and the first
-campaign-table execution slice exists.
-Work only on the lead filter branch. Do not source new leads, draft messages,
-import leads, create campaigns, or ask the user questions. Your only live
-campaign mutation is calling `save_rubrics` after the production rubrics are
-ready.
-
-Required inputs:
-
-- `campaignId`
-- campaign revision or `campaignUpdatedAt`
-- campaign brief content
-- selected source decision and provider/list state
-- `selectedLeadListId`
-- `workflowTableId`
-- initial campaign-table execution slice rows, including row ids/hash when available
-- filter choice
-
-Required first steps:
-
-1. Verify the campaign/source/table ids from the parent thread match the live
-   campaign context.
-2. Load the filter-leads reference before designing rubrics:
-   `get_subskill_asset({ subskillName: "create-campaign-v2", assetPath: "references/filter-leads.md" })`.
-3. Treat campaign state and the campaign-table execution slice as the input of record.
-   Do not require or hunt for local markdown/json artifacts.
-
-Owned outputs:
-
-- Durable campaign rubrics via `save_rubrics({ campaignOfferId, leadScoringRubrics })`
-  when the filter is confirmed and production-shaped rubrics are safe to write.
-  `save_rubrics` is the durable writer.
-- concise filter/rubric summary returned to the parent thread
-
-Do not write or modify local markdown/json artifacts. Durable output is only
-via `save_rubrics` plus the parent-thread summary.
-
-Process:
-
-1. Preserve the approved source decision, source math, and campaign-table slice
-   evidence supplied by the parent; do not re-run sourcing.
-2. Turn the slice's good-fit and false-positive patterns into a strict but
-   campaign-native filter.
-3. Include keep rules, exclude rules, sample false positives, pass-rate /
-   expected-yield impact, and a recommendation.
-4. Add an explicit ability-to-pay or economic-capacity gate unless the brief
-   clearly says the offer is free or has no meaningful budget requirement.
-5. Keep source mechanics out of production rubrics. Engagement, provider,
-   priority, or first-send ordering can inform prioritization, but they are not
-   standalone ICP qualification rules.
-6. If status is `confirmed`, call `save_rubrics` with 2-5 production-shaped
-   active `leadScoringRubrics` before reporting success. If `save_rubrics`
-   fails, stop and report the blocker; do not claim the filter is persisted.
-
-Return a concise final status with:
-
-- filter status: `confirmed`, `confirm-with-user`, or `revise-find-leads`
-- whether `save_rubrics` succeeded and how many active rubrics were persisted
-- strongest keep rules
-- strongest exclusion rules
-- expected pass-rate / yield impact
-- any blocker that prevents message review from joining
-
-Quality bar:
-
-- Every passing lead should be someone the user would be glad to hear back
-  from.
-- Do not loosen the filter just to preserve volume.
-- Do not make the filter so narrow that it contradicts the approved source
-  unless the sample evidence clearly requires it.