Product Feature Research
wiki / competitors / Multica / features / Multica - Reusable Skills

Multica — Reusable Skills

What it does

Captures team-level know-how as codified, executable skill definitions that any agent on the workspace can use. "Write a skill once, and every agent on your team can use it. Your skill library compounds over time."

Schema (from captured SKILL.md example)

SKILL: Write Migration

DESCRIPTION:
  Generate a SQL migration file based on the requested schema changes.
  Validates against the current database state and generates both up
  and down migrations.

STEPS:
  1. Analyze the current schema from migrations/
  2. Generate migration SQL with proper ordering
  3. Validate with sqlc compile
  4. Run tests against a fresh database

So a Skill is essentially a named, parameterizable runbook — markdown frontmatter + ordered steps + reference files. The agent reads the skill and executes the steps using its own toolchain.

Target ICP

Engineering teams that have repeatable operational tasks (deploys, migrations, code review, CI fixes, postmortems) and want them to compound across agents and team members.

How it works (claimed)

"Day 1: you teach an agent to deploy. Day 30: every agent deploys, writes tests, and does code review. Your team's capabilities grow exponentially."

Skills sit in a workspace-level library. Each agent on assignment can invoke applicable skills. "Package knowledge into skills that any agent can execute. Deploy to staging, write migrations, review PRs — all codified."

Differentiators

  1. Vendor-neutral execution — same skill works whether the agent is Claude Code, Codex, or Cursor, because the skill is markdown + steps, not provider-specific code.
  2. Team library, not per-user — explicit "team-wide sharing" framing.
  3. Compound growth — pitched as the lever that makes agent-augmented teams produce non-linear output.

Versus the rest of the vault

This is Multica-only in the captured set. Replit has Connectors and Anything has Custom Instructions, but neither is a team-shared, executable runbook library the way Multica's Skills are.

💭

For CreateOS, the analog is "named agent presets" (research agent / trading agent / support agent — pre-wired to MPP + router) — see CreateOS - Positioning and Wedge and the Differentiate-on-these-5 list in wiki/playbook/04 - UX Recommendations. Multica is doing the team-internal version; CreateOS could do the marketplace-distributed version with revenue share to skill authors.

Sources