docs(planning): cartridge catalogue 2026-06-01 — 941-candidate plan (#179)

## Summary

Single planning artefact: a 941-candidate cartridge catalogue plan
synthesised from 6 parallel Explore-subagent reports run today. Three
deliverables in one document:

1. **Framework** — three orthogonal axes (**substrate × tier × trust**)
so the catalogue scales to ~1000 by construction.
2. **Three decision points** for owner before catalogue expansion:
- **D1**: one-cartridge-multi-protocol vs sibling-family naming (\`aws\`
w/ \`protocols: [MCP, LSP, DAP, BSP]\` vs \`aws-mcp\` + \`aws-lsp\` + …
— affects inventory shape by ~90 cartridges in dev-tools alone)
- **D2**: retire-vs-rewrite the three stub tools (provisioner /
configurator / panel-harness — Agent C found these are
README-and-stub-only since 2026-04-21)
- **D3**: backfill the 14 canonical-only LSPs first (\`cloud-lsp\`,
\`container-lsp\`, \`database-lsp\`, \`git-lsp\`, \`iac-lsp\`,
\`k8s-lsp\`, \`librarian-mcp\`, \`npc-mcp\`, \`observe-lsp\`,
\`proof-lsp\`, \`queues-lsp\`, \`secrets-lsp\`, \`ssg-lsp\`,
\`stack-orchestrator-mcp\`)
3. **Populated catalogue** — 941 candidate cartridges across 6 substrate
buckets:
   - §1 Data & Storage: 119
   - §2 Dev Tools & Languages: 163 *(this slice is gated on D1)*
   - §3 Cloud, Infra, Container: 150
- §4 Observability, Security, Governance: 202 *(80% Teranga/Shield by
trust placement)*
   - §5 Domain, IO, Hardware, Human: 127
   - §6 AI, ML, Agentic, Knowledge: 180

Each candidate tagged with effort (S/M/L) and trust-tier guess
(Ayo/Eno/Teranga/Shield). ~50-80 dedupe collisions across slices; unique
count ~850-900.

Plus a **4-phase prioritisation** (Phase 0 tooling fix → Phase 1
backfill → Phase 2 high-leverage waves → Phase 3 depth fills → Phase 4
exotic) and **5 open questions** for owner that gate any Phase
converting to PR-ready work.

## Why now

This complements the existing planning trail:
- \`hyperpolymath/standards\` BoJ capability-expansion epic
- \`hyperpolymath/boj-server\` #100 (vector-DB + RAG wave — tracked, not
yet built)
- \`hyperpolymath/boj-server\` #101 (multimodal wave — tracked, not yet
built)
- \`hyperpolymath/boj-server\` #87 (the 13-item capability-expansion
roadmap epic)

…by giving us a structured frame within which all three planned waves
slot, rather than each being a one-off issue.

## Joshua's area

The catalogue explicitly excludes any cartridge that would collide with
Joshua's cartridge-fetcher WIP (PRs #169 + #171). Candidates in §6
around \"agentic frameworks + workflow engines\" stay neutral on what
\`agent-mcp\` / \`local-coord-mcp\` already cover.

## Provenance

- 6 Explore subagents fanned out at ~12:30Z 2026-06-01; all returned
within ~30 min.
- Existing 125-cartridge inventory + 14-cartridge canonical-only list
cross-referenced against every candidate.
- Subagent transcripts at
\`/tmp/claude-1000/-home-hyperpolymath-developer-repos/.../tasks/\`.

## Test plan

- [x] Pre-commit hook accepts the file (SPDX + Copyright at top)
- [x] Pure documentation — no code changes
- [ ] CI passes (governance / dogfood-gate / scorecards)

## Refs

Refs #87, #100, #101.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: hyperpolymath