Voice & knowledge

Profile-wide Core Voice, Licensed Voice, inspiration, facts, and boundaries

Voice & knowledge is the active Synfluencer's persistent context library. Add writing, email, documents, webpages, transcripts, recordings, or video once, then control whether the material may become active Core Voice, a permissioned Licensed Voice candidate, abstract Style Inspiration, reusable Knowledge, or remain an inactive library item.

Why it is separate from From Source

• Voice Library is global to the profile. Core Voice is the active, authoritative writing context used for drafts. Licensed Voice holds permissioned candidates by another author. Eligible candidates move into Core only after promotion, with authorship and attribution preserved.
• Style Inspiration is influence, not identity. Synfluencer retrieves analyzed hook, structure, register, and payoff mechanisms. The original post wording, author identity, claims, and anecdotes are not sent to generation.
• Knowledge can be automatic or draft-specific. Turn on Use when relevant for facts and guidance that may apply broadly, or select a library item from From Source's Knowledge tab when only one draft needs it.
• From Source is temporary composition context. Its pasted text, links, files, and selected Knowledge items stay attached to that draft in Review; they do not silently change the profile.

Add approved voice evidence

1. Open Voice & knowledge below the profile selector in the sidebar.
2. Choose Voice Library, then choose Core Voice or Licensed Voice. Paste text, add a public URL, upload a supported file, import files, or fetch public posts from one or more X accounts.
3. Identify authorship and confirm you have permission to use the material.
4. For mixed-speaker recordings, confirm the profile speaker or select only the relevant transcript passages.
5. For third-party material, confirm permission and add it to Licensed Voice; public availability alone is not permission. After analysis, promote an eligible item or batch to Core Voice before it can shape drafts.
6. Select any pasted writing item later to review and edit its full title and content.

Use Style Inspiration without copying a creator

1. Open Style Inspiration and add permissioned examples directly, from files, or from public X accounts.
2. Review the inactive analysis. It extracts mechanism-level patterns such as opening type, structure, emotional register, and payoff—not reusable source wording.
3. Activate eligible patterns. Future generation combines those mechanisms with this Synfluencer's Core/Licensed Voice, Persona/Brand, intent, source, and destination constraints.
4. Deactivate or restore a prior activation whenever the influence no longer fits.

Maintain reusable knowledge

Use Knowledge for product facts, positioning, terminology, customer context, approved claims, legal boundaries, and other material that should keep future drafts accurate. Review old items, disable automatic reuse when guidance expires, and delete material that should no longer be available.

Analyze a large import before it changes anything

1. Choose X accounts for up to 10 handles/profile URLs and 100 public posts per account, or Import files for up to 1,000 JSON, JSONL, CSV, text, or Markdown items. Replies and reposts start off.
2. Review the item count, authorship, permission state, provider disclosure, and maximum analysis charge.
3. Synfluencer scores standalone clarity, voice/profile fit, hook, payoff, structure, originality, account-relative performance, duplicates, intent/Shape affinity, and retrieval usefulness.
4. Open the inactive report to compare source accounts and ranked items. Quality anchors/approved/review/excluded describe writing usefulness; Activation excluded separately describes permission and authorship eligibility.
5. Activation is a separate confirmation. Direct Core Voice requires self-authored or selected mixed-author material. Licensed Voice requires explicit permission, and only its eligible examples are promoted into Core; review/excluded candidates remain Licensed and inactive. Inspiration requires permission and retrieves mechanisms only.
6. Choose Blend to promote eligible examples alongside the active Core set or Replace to deactivate the current Core set without deletion. Restore previous library demotes the promoted examples and restores the exact prior activation.

Navigate a large library

Core and Licensed controls show active and candidate counts. Each Voice card is labeled Active Core, Eligible inactive, Needs review, Excluded, or Inactive so processing status cannot be mistaken for use in drafts. The library shows six items at a time on desktop and mobile. Search by name or filter by file type and upload date; the visible range and page controls always reflect every matching stored source.

Privacy and retrieval

The library can grow without a five-item cap, but each generation receives only a small, relevant subset. Source bodies are editor-only. Audio and video use the configured zero-data-retention transcription routes. Batch analysis names the configured analysis and embedding-provider transfer before it begins, uses a no-data-collection route, and leaves every derived vector inactive until the report passes the separate activation boundary. Inspiration vectors contain only abstract patterns, while Voice and Knowledge vectors remain permission- and approval-gated. Source bodies never enter task, generation-run, or workspace event metadata.