kb-query
Skill2 starsChanged 9 days agoMITView on GitHub
Read-only query of the existing study knowledge base, persisting a query-session (without writing any vault file). Use when the user asks "what does the KB say about X / search my wiki / what was that Y I learned / has Z ever been covered". Read-only; to keep a conclusion in the wiki, follow up explicitly with kb-save.
6 other projects use a version of this.
Adopted by
Most recently taken up by
- janezslapnik/llm-wiki-kbtheir own version, first seen Nov 1, 2026
- walkwithedi/drive-wikitheir own version, first seen Nov 1, 2026
- walkwithedi/evergreen-wikitheir own version, first seen Oct 11, 2026
- davidmatousek/tachitheir own version, first seen Oct 10, 2026
- viniciusbedum/vb-ai-contexttheir own version, first seen Sep 25, 2026
- tjkuhns/explodabletheir own version, first seen Apr 19, 2026
Used with
Projects with it also use kb-ingest.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
shaowen310Skillkb-ingest
Thin orchestrator that turns PowerPoint `.pptx` slides into knowledge-base-ready, slide-level JSONL chunks for RAG / agent pipelines. Delegates extraction to the pptx2md skill, then parses the Markdown per the deterministic output contract (slide anchors, field extraction, image/encoding rules) into one record per slide with title, body, images, and metadata. Triggers on: "把PPT做成知识库", "PPT转知识库给AI用", "ingest PPTX into KB", "build RAG dataset from slides", "把幻灯片切片入库". Best for building agent/RAG knowledge bases, search indexes, or training corpora from slide decks. Do NOT use for translating PPTX (use the pptx-translate skill) or for plain markdown export without chunking (use the pptx2md skill).
Also in 4 projects that use this
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- davidmatousek/tachi★ 95same name
- viniciusbedum/vb-ai-context★ 2same name
- walkwithedi/evergreen-wiki★ 2same name
- janezslapnik/llm-wiki-kb★ 1same name
- tjkuhns/explodable★ 1same name
- walkwithedi/drive-wiki★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- ingestSkill
- kb-postmortemSkill
- kb-qaSkill
- kb-querySkill
- kb-reviewSkill
- kb-saveSkill
- pipeline-doctorSkill
- skill-evolveSkill
- source-preflightSkill
- source-xraySkill
- wiki-lint-semanticSkill
- ingestSkill
- kb-postmortemSkill
- kb-qaSkill
- kb-reviewSkill
- kb-saveSkill
- pipeline-doctorSkill
- skill-evolveSkill
- source-preflightSkill
- source-xraySkill
- wiki-lint-semanticSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 0 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 20
Read: wiki/index.generated.md, wiki/concepts/_registry.yaml, and the relevant concept/topic/comparison/synthesis/source/lesson pages.
- Very specific: line 21
If the answer might be worth saving, read docs/skill-runtime/save-back-policy.md to decide whether to emit a save candidate.
- Very specific: line 51
Run check-session after writing the query-session.
- Very specific: line 71
No change under wiki/.
“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
kb-query — read-only query + query-session persistence
Answer the user's questions about existing KB content. Read-only: it writes no file under wiki/, but
it must persist a query-session for later kb-save and for audit. The execution layer is
scripts/pipeline.py; this skill only orchestrates, surfaces acceptance, and constrains artifacts.
1. Triggers / Non-triggers
- Triggers: "what does the KB say about X", "search my wiki", "what was that Y I learned", "has Z ever been covered".
- Non-triggers: ingesting a new source (use
ingest); writing a query result back (usekb-save); working the Review-Queue (usekb-review); a semantic health check (usewiki-lint-semantic); plain summary/translate/explain of external text (a normal answer — no query-session unless the user explicitly queries the wiki).
2. Inputs
- The user's question; optional domain, concept name, source name, page path, or time range.
- Read:
wiki/index.generated.md,wiki/concepts/_registry.yaml, and the relevant concept/topic/comparison/synthesis/source/lesson pages. - If the answer might be worth saving, read
docs/skill-runtime/save-back-policy.mdto decide whether to emit a save candidate.
3. Outputs
- An in-chat answer citing the relevant vault pages (wikilinks) and source location (source §section / page).
- The query-session files under
pipeline-workspace/query-sessions/<run_id>/. - It does not write
wiki/, does not touchlog.md, does not promote, does not create proposed pages.
4. Dependencies
- Protocols:
docs/skill-runtime/save-back-policy.md(save-candidate admission),docs/skill-runtime/schema.md(page types). - CLI:
scripts/pipeline.py check-session --id <run_id>to self-check the query-session. - Saving is handed to
kb-saveonly; this skill does not inline save logic.
5. Persisted artifacts
Under pipeline-workspace/query-sessions/<run_id>/:
question.md: the original question.answer.md: this answer.related_pages.json: the list of vault page paths involved.candidate_write_set.json: pages worth saving (synthesis/comparison/learning-path candidates), else[].evidence_refs.json:[{"source": "...", "sections": ["..."]}];[]if no source evidence.
6. CLI commands
python scripts/pipeline.py check-session --id <run_id>Run check-session after writing the query-session; on failure, fix the session files and do not treat a failed session as a save basis.
7. Workflow
| Sub-unit | Input | Output | Acceptance | Persisted | Failure stop |
| Q1 locate material | question + index/registry | relevant page list | paths exist, prefer published content | related_pages.json draft | vault/index missing |
| Q2 answer | relevant page bodies | answer with wikilinks / source location | no source-less claims; separate established vs inferred | answer.md draft | say so if evidence is thin |
| Q3 save-candidate judgement | answer + save-back-policy | candidate_write_set/evidence_refs | plain explanation / one-off fact → [] | JSON files | empty if candidate has no evidence |
| Q4 session self-check | session dir | check-session result | check-session passes | full query-session dir | self-check fails |
8. Failure stops / recovery
wiki/index.generated.md or registry missing; relevant pages missing; query-session write fails;
check-session fails; the user asks to write the wiki mid-query (stop this skill, switch to kb-save);
never emit a save candidate without source evidence. Recovery: re-run check-session after fixing the
session files; the session dir is the durable handoff to kb-save.
9. Acceptance criteria
- No change under
wiki/. - The five query-session files exist and the JSON parses.
- Paths in
related_pages.jsonreally exist. candidate_write_set.jsonis[]for plain explanation / translation / one-off facts.python scripts/pipeline.py check-session --id <run_id>passes.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

