@sellable/mcp 0.1.158 → 0.1.159

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -151,17 +151,17 @@ unless the user asks for them.
151
151
  Good opening:
152
152
 
153
153
  ```text
154
- I’ll help you launch this as a Sellable campaign. First I’ll confirm who we’re
155
- sending from and which company this is for, then I’ll turn that into a campaign
156
- brief before lead sourcing or anything can send.
154
+ I’ll help you launch this as a Sellable campaign. First I’ll research the
155
+ person/company this is for, then I’ll turn that into a campaign brief before
156
+ lead sourcing or anything can send.
157
157
  ```
158
158
 
159
159
  Good identity setup:
160
160
 
161
161
  ```text
162
- I’ll confirm the client/company this campaign is for before we pick the target,
163
- offer, proof, and lead source. Paste a LinkedIn profile or company website and
164
- I’ll resolve the campaign identity before the setup choices.
162
+ I’ll ask for the LinkedIn profile URL first, with the company website as a
163
+ fallback. Then I’ll show the researched recommendation and you can correct
164
+ anything stale before we pick the target, offer, proof, and lead source.
165
165
  ```
166
166
 
167
167
  Bad:
@@ -243,8 +243,9 @@ For LinkedIn engagement, the second gate is not "approve source" in the abstract
243
243
  It is a concrete scrape approval: show a compact `## Source Recommendation`
244
244
  with the ~300 headline-fit prospect goal, source sample assumption, selected-post
245
245
  table, total visible pool, estimated headline-fit pool, 10% planning floor,
246
- first-pass campaign setup, and fallback. The approval question should be "Approve
247
- scraping N selected LinkedIn posts?"
246
+ first-pass campaign setup, and fallback. The approval question should be
247
+ "Approve scraping N recommended LinkedIn posts?" where N comes from the
248
+ source-candidate math, not the default promoted sample count.
248
249
 
249
250
  That scrape approval is single-use. Once the user approves it, do not replay
250
251
  the source card or ask for the same approval again. Acknowledge the approval in