Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/kyopark2014/agent-plugins/knowledge-synthesisnpx skills add kyopark2014/agent-plugins --skill knowledge-synthesisgit clone --depth 1 https://github.com/kyopark2014/agent-pluginsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00038 | $0.01831 |
| Opus 5 | $0.00019 | $0.00915 |
| Sonnet 5 | $0.00008 | $0.00366 |
| Haiku 4.5 | $0.00004 | $0.00183 |
Grade A, and why
knowledge-synthesis scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
94% identical to knowledge-synthesis — 1 line differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 258 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge Synthesis
The last mile of enterprise search. Takes raw results from multiple sources and produces a coherent, trustworthy answer.
The Goal
Transform this:
~~chat result: "Sarah said in #eng: 'let's go with REST, GraphQL is overkill for our use case'"
~~email result: "Subject: API Decision — Sarah's email confirming REST approach with rationale"
~~cloud storage result: "API Design Doc v3 — updated section 2 to reflect REST decision"
~~project tracker result: "Task: Finalize API approach — marked complete by Sarah"
Into this:
The team decided to go with REST over GraphQL for the API redesign. Sarah made the
call, noting that GraphQL was overkill for the current use case. This was discussed
in #engineering on Tuesday, confirmed via email Wednesday, and the design doc has
been updated to reflect the decision. The related ~~project tracker task is marked complete.
Sources:
- ~~chat: #engineering thread (Jan 14)
- ~~email: "API Decision" from Sarah (Jan 15)
- ~~cloud storage: "API Design Doc v3" (updated Jan 15)
- ~~project tracker: "Finalize API approach" (completed Jan 15)
Deduplication
Cross-Source Deduplication
The same information often appears in multiple places. Identify and merge duplicates:
Signals that results are about the same thing:
- Same or very similar text content
- Same author/sender
- Timestamps within a short window (same day or adjacent days)
- References to the same entity (project name, document, decision)
- One source references another ("as discussed in ~~chat", "per the email", "see the doc")
How to merge:
- Combine into a single narrative item
- Cite all sources where it appeared
- Use the most complete version as the primary text
- Add unique details from each source
Deduplication Priority
When the same information exists in multiple sources, prefer:
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)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 258 lines · 38 tokens per session scan A 930ea22079de
knowledge-synthesis is a skill published in the GitHub repository kyopark2014/agent-plugins (4 stars, last pushed 25d ago), licensed MIT. It adds 38 tokens to every session and 1,831 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 94% identical to knowledge-synthesis, differing in 1 line, and is treated as a copy.
Other skills, from other repositories
ask-zilliz
Zilliz Cloud onboarding and usage assistant. Helps users understand Zilliz Cloud, choose the right plan, estimate costs, write code, debug issues, and adopt new features like Functions, Volumes, and Global Clusters. Use this skill whenever the user asks about Zilliz Cloud — including plan selection, pricing, cost…
diagnose
Use when the user reports that a Zilliz Cloud cluster or Milvus collection is unhealthy, slow, stuck, returning errors, hitting quotas, or otherwise misbehaving — or when they ask "what's wrong with...", "why is ... slow", "diagnose ...", "troubleshoot ...".
collection
Use when the user wants to create, list, describe, drop, rename, load, release, or manage collections and collection aliases in Milvus.
cluster
Use when the user wants to create, list, describe, delete, suspend, resume, or modify Zilliz Cloud clusters.
external-collection
Use when the user wants to trigger, describe, or list refresh jobs for an external collection (a collection backed by an external data source such as Vector Lake). Note this is the data-plane refresh workflow, not collection CRUD -- for create/drop/load see the collection skill.
backup
Use when the user wants to create, list, describe, delete, export, or restore backups, or manage backup policies on Zilliz Cloud.