skills/explore-knowledge/SKILL.md
Use whenever the user asks about this project's decisions, architecture, conventions, or how something works in this codebase — do NOT answer from general knowledge, always look it up in docs/. Trigger on questions like "why did we choose X?", "how does Y work here?", "what's our convention for Z?", "what docs reference X?", or any question about this project's specific choices. Also use for explicit /autology:explore-knowledge commands (overview, neighborhood, path). Searches docs/, follows wikilinks, and synthesizes grounded answers from the knowledge base.
npx skillsauth add curt-park/autology explore-knowledgeInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Answer user questions about the project by traversing the autology knowledge graph. Search docs/ for relevant nodes, follow wikilinks to build context, and return rich answers grounded in documented decisions, conventions, and architecture.
Use explore instead of reading a doc directly when: the user asks a question (rather than requesting a specific doc), or when the topic may span multiple connected nodes.
When the user asks a question about the project (conventions, architecture, decisions, rationale):
docs/ for keywords from the questionThe <node> argument is a title-slug — the filename without the .md extension (e.g., redis-storage-decision for docs/redis-storage-decision.md).
/autology:explore-knowledge overview)Process:
docs/*.md, read each file's frontmatter and wikilinksOutput:
/autology:explore-knowledge <node>)Process:
[[slug]]docs/ for [[target-slug]] to find nodes that reference the targetOutput: node title, type, tags, and connections for each hop
/autology:explore-knowledge path A B)Process:
Output: A → [intermediate] → B with each hop labeled
/autology:explore-knowledge overview # graph overview
/autology:explore-knowledge <node> # neighborhood (2-hop BFS)
/autology:explore-knowledge path A B # shortest path
| Mistake | Fix |
|---------|-----|
| Answer from memory instead of docs | Always Read the actual doc nodes before answering |
| Stop at 1-hop neighbors | Follow wikilinks to 2-hop for richer context |
| Ignore node type and tags | Include type/tags in answers — they add classification context |
| Only follow outgoing links in neighborhood | Also grep docs/ for [[node-name]] to find nodes that link TO the target |
| Speculate when question has false premise | State what IS documented; never construct a rationale not found in docs |
development
Use after significant actions (commits, decisions, refactors) or when asked to triage changes — classifies knowledge items as existing or new and provides topology hints for sync and capture.
development
Use when existing autology docs/ nodes need syncing or updating — when a doc is out of date after a code change, after refactors, when triage identifies existing nodes to verify, or for periodic full audits. Not for new items (use capture) or knowledge questions (use explore).
testing
Use to permanently record a project decision, convention, pattern, or known issue into the knowledge base (docs/) — triggers on explicit save intent ("remember this", "save this", "document this"), decision or convention announcements ("we decided", "settled on", "the rule is"), or triage output classifying new items to capture. Distinct from conversational memory — this writes to docs/.
development
--- name: autology-workflow description: Use after significant project actions — commit, push, PR — or when the team settles on how to build or operate the system: a technology choice, a process, a rule, or a standard. Running close to the action keeps captured knowledge more accurate and complete. --- ## Overview Route to the right autology skill after significant actions. Knowledge captured close to the action is more accurate and complete — this workflow keeps docs/ in sync with what actual