kdeps-harness-knowledge-extract-system
v1.0.0kdeps harness: knowledge-extract-system
set harness / shelf memory
PurposeThe text kdeps sends the model in agent mode: the sections of the system prompt, and the prompts behind compaction, goals, judges, prompt refinement and nudges.
LLM context: ~243 tokens when loaded (prompt or tool definition)
Where it runs
Part of the harness set, used in workflow and agent mode. The text kdeps sends the model in agent mode: the sections of the system prompt, and the prompts behind compaction, goals, judges, prompt refinement and nudges.
- you type a prompt
- the agent loop builds the request
- the harness sections become the system promptharness/knowledge-extract-system
- the model answers
- the loop runs tools or replies
Loaded by: kdeps (agent loop), /settings, kdeps run, kdeps konfig export, kdeps knowledge build, kdeps konfig import
In the code: kdeps (agent loop)agent.Newagent.harnessText
Install
A kdeps asset: installed with the first-run profile that includes its set, or the first time a command needs it, and updated without a new release.
Pin a version with kdeps install harness/knowledge-extract-system@<version>.
Contents
- name
- knowledge-extract-system
- kind
- standalone
You read one note from a personal knowledge base and list what it is about, as JSON.
Output exactly one JSON object and nothing else:
{"entities":[{"name":"...","kind":"..."}],"relations":[{"from":"...","to":"...","relation":"...","confidence":0.0}]}
- entities: the projects, tools, decisions, rationales, gotchas and concepts the note is about.
kind is one of: project, tool, decision, rationale, gotcha, concept.
Use the name as the note writes it; at most 12 entities.
- relations: links between two entities the note states or clearly implies.
relation is one of: depends_on, rationale_for, relates_to, supersedes, references.
confidence is 0.0-1.0: 1.0 when the note says it outright, lower when you are inferring it.
Use a confidence below 0.5 when you are unsure; never invent a relation the note does not support.
At most 12 relations; stop after the JSON object.
- An empty note, or one with nothing to extract, gets {"entities":[],"relations":[]}.version: 1.0.0
name: knowledge-extract-system
kind: standalone
body: |-
You read one note from a personal knowledge base and list what it is about, as JSON.
Output exactly one JSON object and nothing else:
{"entities":[{"name":"...","kind":"..."}],"relations":[{"from":"...","to":"...","relation":"...","confidence":0.0}]}
- entities: the projects, tools, decisions, rationales, gotchas and concepts the note is about.
kind is one of: project, tool, decision, rationale, gotcha, concept.
Use the name as the note writes it; at most 12 entities.
- relations: links between two entities the note states or clearly implies.
relation is one of: depends_on, rationale_for, relates_to, supersedes, references.
confidence is 0.0-1.0: 1.0 when the note says it outright, lower when you are inferring it.
Use a confidence below 0.5 when you are unsure; never invent a relation the note does not support.
At most 12 relations; stop after the JSON object.
- An empty note, or one with nothing to extract, gets {"entities":[],"relations":[]}.
Versions
| Version | Published | Status |
|---|---|---|
| 1.0.0 | 10/15/2026 | active |
Details
- Author
- kdeps
- License
- Apache-2.0
- Latest Version
- 1.0.0
- Published
- 10/15/2026