norandom/mmhmem
Save token money on scale. The missing economic layer for AI cost optimization. mmhmem instead of "oh my god".
Copy this rig
$ git clone --depth 1 https://github.com/norandom/mmhmem MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit norandom/mmhmem/.opencode/skills/kiro-debug .claude/skills/kiro-debug $ npx degit norandom/mmhmem/.opencode/skills/kiro-discovery .claude/skills/kiro-discovery $ npx degit norandom/mmhmem/.opencode/skills/kiro-impl .claude/skills/kiro-impl $ npx degit norandom/mmhmem/.opencode/skills/kiro-review .claude/skills/kiro-review $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-batch .claude/skills/kiro-spec-batch $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-design .claude/skills/kiro-spec-design $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-init .claude/skills/kiro-spec-init $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-quick .claude/skills/kiro-spec-quick $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-requirements .claude/skills/kiro-spec-requirements $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-status .claude/skills/kiro-spec-status $ npx degit norandom/mmhmem/.opencode/skills/kiro-spec-tasks .claude/skills/kiro-spec-tasks $ npx degit norandom/mmhmem/.opencode/skills/kiro-steering-custom .claude/skills/kiro-steering-custom $ npx degit norandom/mmhmem/.opencode/skills/kiro-steering .claude/skills/kiro-steering $ npx degit norandom/mmhmem/.opencode/skills/kiro-validate-design .claude/skills/kiro-validate-design $ npx degit norandom/mmhmem/.opencode/skills/kiro-validate-gap .claude/skills/kiro-validate-gap $ npx degit norandom/mmhmem/.opencode/skills/kiro-validate-impl .claude/skills/kiro-validate-impl $ npx degit norandom/mmhmem/.opencode/skills/kiro-verify-completion .claude/skills/kiro-verify-completion $ npx degit norandom/mmhmem/.opencode/skills/lesscode-discovery .claude/skills/lesscode-discovery $ npx degit norandom/mmhmem/deploy/codex/skills/mmhmem-recap .claude/skills/mmhmem-recap $ npx degit norandom/mmhmem/deploy/skills/unified-memory .claude/skills/unified-memory
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Skills (20)
Kiro (4 steering · 12 specs)
Specs: agent-tool-routing, ci-build-time, ci-dependency-cache, client-usage-stats, corpus-warmup-prune, cross-agent-memory-server, install-one-liner, local-mode, proactive-session-checkpoints, read-routing-hooks, reliable-installers, static-linux-release (loaded on demand, not counted in context tax)
How Kiro files are scored: methodology.