npx skills add ...
npx skills add anthropics/knowledge-work-plugins --skill knowledge-synthesis
npx skills add anthropics/knowledge-work-plugins --skill knowledge-synthesis
Combines search results from multiple sources into coherent, deduplicated answers with source attribution. Handles confidence scoring based on freshness and authority, and summarizes large result sets effectively.
The last mile of enterprise search. Takes raw results from multiple sources and produces a coherent, trustworthy answer.
Transform this:
Into this:
The same information often appears in multiple places. Identify and merge duplicates:
Signals that results are about the same thing:
How to merge:
When the same information exists in multiple sources, prefer:
Keep as separate items when:
Every claim in the synthesized answer must be attributable to a source.
Inline for direct references:
Source list at the end for completeness:
Not all results are equally trustworthy. Assess confidence based on:
| Recency | Confidence impact |
|---|---|
| Today / yesterday | High confidence for current state |
| This week | Good confidence |
| This month | Moderate — things may have changed |
| Older than a month | Lower confidence — flag as potentially outdated |
For status queries, heavily weight freshness. For policy/factual queries, freshness matters less.
| Source type | Authority level |
|---|---|
| Official wiki / knowledge base | Highest — curated, maintained |
| Shared documents (final versions) | High — intentionally published |
| Email announcements | High — formal communication |
| Meeting notes | Moderate-high — may be incomplete |
| Chat messages (thread conclusions) | Moderate — informal but real-time |
| Chat messages (mid-thread) | Lower — may not reflect final position |
| Draft documents | Low — not finalized |
| Task comments | Contextual — depends on commenter |
When confidence is high (multiple fresh, authoritative sources agree):
When confidence is moderate (single source or somewhat dated):
When confidence is low (old data, informal source, or conflicting signals):
When sources disagree:
Always surface conflicts rather than silently picking one version.
Present each result with context. No summarization needed — give the user everything:
Group by theme and summarize each group:
Provide a high-level synthesis with the option to drill down:
Do not:
Do:
1. The most complete version (fullest context)
2. The most authoritative source (official doc > chat)
3. The most recent version (latest update wins for evolving info)Sarah confirmed the REST approach in her email on Wednesday.
The design doc was updated to reflect this (~~cloud storage: "API Design Doc v3").Sources:
- ~~chat: #engineering discussion (Jan 14) — initial decision thread
- ~~email: "API Decision" from Sarah Chen (Jan 15) — formal confirmation
- ~~cloud storage: "API Design Doc v3" last modified Jan 15 — updated specificationThe team decided to use REST for the API redesign. [direct statement]Based on the discussion in #engineering last month, the team was leaning
toward REST for the API redesign. This may have evolved since then.I found a reference to an API migration discussion from three months ago
in ~~chat, but I couldn't find a formal decision document. The information
may be outdated. You might want to check with the team for current status.I found conflicting information about the API approach:
- The ~~chat discussion on Jan 10 suggested GraphQL
- But Sarah's email on Jan 15 confirmed REST
- The design doc (updated Jan 15) reflects REST
The most recent sources indicate REST was the final decision,
but the earlier ~~chat discussion explored GraphQL first.[Direct answer synthesized from results]
[Detail from source 1]
[Detail from source 2]
Sources: [full attribution][Overall answer]
Theme 1: [summary of related results]
Theme 2: [summary of related results]
Key sources: [top 3-5 most relevant sources]
Full results: [count] items found across [sources][Overall answer based on most relevant results]
Summary:
- [Key finding 1] (supported by N sources)
- [Key finding 2] (supported by N sources)
- [Key finding 3] (supported by N sources)
Top sources:
- [Most authoritative/relevant source]
- [Second most relevant]
- [Third most relevant]
Found [total count] results across [source list].
Want me to dig deeper into any specific aspect?[Raw results from all sources]
↓
[1. Deduplicate — merge same info from different sources]
↓
[2. Cluster — group related results by theme/topic]
↓
[3. Rank — order clusters and items by relevance to query]
↓
[4. Assess confidence — freshness × authority × agreement]
↓
[5. Synthesize — produce narrative answer with attribution]
↓
[6. Format — choose appropriate detail level for result count]
↓
[Coherent answer with sources]