Knowledge Stack
- 1 installs
- 28 repo stars
- Updated June 29, 2026
- nickcrew/claude-cortex
Describes where a specific developer's durable knowledge lives across repos and a memory vault and which layer to consult for a given question.
About
Maps where the author's durable knowledge lives (repo backlog docs and a Basic Memory vault) and which layer to consult for architecture, product, or personal/career questions. A developer uses it to route to authoritative artifacts instead of lossy stored-memory summaries when working in the author's repos.
- Directs reasoning to backlog docs and the memory vault over lossy summaries
- Author- and repo-specific; not a general-purpose skill
Knowledge Stack by the numbers
- 1 all-time installs (skills.sh)
- Ranked #2,473 of 3,280 Productivity & Planning skills by installs in the Skillselion catalog
- Data as of Aug 4, 2026 (Skillselion catalog sync)
npx skills add https://github.com/nickcrew/claude-cortex --skill knowledge-stackAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1 |
|---|---|
| repo stars | ★ 28 |
| Last updated | June 29, 2026 |
| Repository | nickcrew/claude-cortex ↗ |
What it does
Describes where a specific developer's durable knowledge lives across repos and a memory vault and which layer to consult for a given question.
Files
Knowledge Stack
Nick's durable knowledge lives in two layers. Stored memory summaries are lossy; these are the real artifacts. Consult them before reasoning.
The Split
Per-repo dev work → backlog docs and backlog tasks inside each repo.
- Architecture, product strategy, implementation plans, active priorities, roadmap notes.
- Consistent pattern across Nick's repos under
~/Developer/. - For CLI mechanics, see the
backlog-mdskill.
Cross-cutting personal and career → Basic Memory vault (via the basic-memory MCP).
- Job search pipeline, interview prep state, tax amendment, foreclosure/tenant situation, Hard Stuff drafts, brand and design work, strategic positioning.
- Obsidian is an optional UI, not required. The vault is plain markdown; any editor works.
Ephemeral conversation context → this chat. Promote to the right layer when worth keeping.
Operational Rules
1. Before reasoning about a specific repo, run backlog doc list in that repo. Pull relevant docs with backlog doc view <id> before forming conclusions from stored memory. 2. Before reasoning about cross-cutting work, search Basic Memory (search_notes, then read_note) before reasoning from stored memory. 3. When a chat surfaces something durable, propose writing it to the right layer. Don't silently drop context. 4. Don't duplicate content across layers. If something could plausibly live in both, ask Nick which layer owns it.
Failure Modes to Avoid
- Reasoning from stored memory summaries when fresher truth exists in backlog docs or the vault — produces stale or wrong answers.
- Writing to Basic Memory what belongs in a repo's backlog docs, or vice versa — produces knowledge drift across layers that compounds over time.
- Treating Obsidian as required — Nick doesn't want the Electron app running. The vault is markdown files; use whatever editor or the MCP directly.
- Enumerating "Nick's repos" from memory and getting it wrong — just
ls ~/Developer/or check the current working directory.
Related skills
Grill MeHave an agent relentlessly pressure-test their plan or design until every decision branch is resolved and shared understanding is reached.691k196kGrill With DocsPressure-test architecture decisions against their live codebase and living documentation before writing production code.585k196k
Lark CalendarLet their coding agents read, create, and manage events inside Lark (Feishu) Calendar directly from natural language instructions.493kLark ImLet their coding agent send instant notifications, post updates, and receive commands directly inside Lark (Feishu) workspaces.493kLark ContactLet their AI coding agents read and write contacts, send messages, and manage user data inside Lark (Feishu).493kLark AttendanceAutomatically fetch and report daily attendance data from Lark (Feishu) into their own tools or agents.493k