Skip to content

/wiki-reflect

Run a self-reflection pass over the whole wiki — look for patterns, gaps, and improvement opportunities.

From plugin
llmwiki
40018 skills18 commands
> /plugin marketplace add Pratiyush/llm-wiki
> /plugin install llmwiki@llmwiki

Command definition

wiki-reflect.md

Run a self-reflection pass over the whole wiki — look for patterns, gaps, and improvement opportunities.

Usage: /wiki-reflect [focus-area]

`$ARGUMENTS` is an optional focus hint: `patterns`, `gaps`, `quality`, `contradictions`, or omit for a general pass.

Unlike `/wiki-lint` (which finds structural issues) or `/wiki-query` (which answers one question), `/wiki-reflect` asks the higher-order question: **what is this wiki trying to say, and what's missing?**

Steps

1. **Read `wiki/index.md` and `wiki/overview.md`** to get the current synthesis.

2. **Read every page under `wiki/sources/`, `wiki/entities/`, and `wiki/concepts/`** — or a representative sample if there are more than 50 pages.

3. **Look for:**

  • **Recurring themes** that span multiple sources — candidates for new concept pages
  • **Implicit comparisons** where two entities or concepts keep appearing together — candidates for `wiki/comparisons/` pages
  • **Unanswered questions** mentioned in multiple sources — candidates for `wiki/questions/` pages
  • **Evolution over time** — how has the user's thinking on a concept changed?
  • **Contradiction clusters** — pages that disagree on the same point
  • **Orphan themes** — interesting ideas mentioned once then never again
  • **Missing entities** — people, tools, or projects mentioned in 3+ sources without their own page

4. **Synthesise a 3-5 paragraph reflection** structured as:

  • **What the wiki knows well** — the most-covered topics, the strongest signal
  • **What the wiki is missing** — gaps that reflection surfaced
  • **Suggested next actions** — new pages, new queries, or new sources to ingest

5. **Ask the user** whether to:

  • Save the reflection as `wiki/syntheses/reflection-<YYYY-MM-DD>.md`
  • Open GitHub issues for any "new page" suggestions
  • Schedule a `/wiki-ingest` of any missing sources the reflection identified

6. **Append to `wiki/log.md`**:

   ## [YYYY-MM-DD] reflect | <short summary of findings>

Focus-area hints

| Hint | What to emphasise | |---|---| | `patterns` | Recurring themes, clusters, evolution over time | | `gaps` | Missing entities/concepts, unanswered questions | | `quality` | Stale summaries, weak writing, duplicated effort | | `contradictions` | Disagreements across pages, unresolved claims | | *(none)* | A general pass — briefly touch each of the above |

Hard rules

1. **Don't edit pages during reflection.** Reflection is read-only. Any changes are proposed to the user, who approves before you touch the wiki. 2. **Distinguish "the wiki says" from "I think".** Cite specific pages (with `[[wikilinks]]`) when describing what the wiki knows. Flag your own interpretations clearly. 3. **Be specific, not vague.** "The wiki covers Python well" is unhelpful. "The wiki has 14 Python sessions, strong coverage of pytest and argparse, but zero on async or typing" is actionable.

Read more
@pratiyush
Ships withllmwiki

LLM-powered knowledge base from your Claude Code, Codex CLI, Cursor, Gemini CLI, and Obsidian sessions. Built on Andrej Karpathy's LLM Wiki pattern.

Stars on GitHub

Momentum

stars in total
400
Star history of Pratiyush/llm-wiki on GitHub, a chart by star-history.com

Other commands on llmwiki.

maintainer
Auto-invoked

maintainer

Command

Meta-skill that loads all llmwiki governance docs and exposes the three maintainer slash commands.

@pratiyush@pratiyushInstall Command
review-pr
Auto-invoked

review-pr

Command

Run the canonical llmwiki code review against a pull request and post findings.

@pratiyush@pratiyushInstall Command
triage-issue
Auto-invoked

triage-issue

Command

Apply labels + milestone + priority to a new issue using the llmwiki triage rules.

@pratiyush@pratiyushInstall Command
wiki-all
Auto-invoked

wiki-all

Command

Run the full llmwiki pipeline end-to-end: build → graph → export all → lint.

@pratiyush@pratiyushInstall Command

More commands from other plugins.