№ 0896Skill
Muse-native ports of community agent plugins
Muse workspace skills ported from a Codex/Claude Code plugin marketplace, translating manifests and hooks into procedures Muse can run with its terminal, browser and files.
# muse-skills Muse-native skills adapted from the [community agent plugins](https://github.com/Israelmusondaayliffe/plugins) marketplace. The plugins repository ships reusable capabilities for Codex, Claude Code, and Claude Cowork. This repository is its Muse equivalent: each plugin ported into a workspace skill for Muse (Meta's personal AI assistant, running on Hatch). ## What changed in adaptation - Codex/Claude plugin manifests (`.codex-plugin`, `.claude-plugin`), host-specific hooks, and agent YAML formats were **not** copied literally. Their substance was translated into procedures Muse can actually run: terminal shell, live browser, filesystem, cron/scheduled work, and subagent delegation. - State paths were re-homed from `~/.claude` / `~/.codex` to `~/workspace/`. - Scheduling surfaces became Hatch cron jobs; checkpoint hooks became explicit checkpoint procedures. - Scripts were reviewed before porting; only safe, useful, stdlib-based helpers were kept. - Nothing user-specific was invented: skills that need personal input (e.g. writing samples for `writing-quality`) ask for it at runtime. ## Skills | Skill | What it does | |---|---| | `agent-ops` | Design, build, and audit reusable agent systems on Hatch. Covers pattern selection (single call, workflows, or autonomous agents), subage... | | `ai-film-studio` | Explicit-only AI film-production planning: concept grill, film brief, production records, shot packets, and iteration supervision. Trigge... | | `brand-studio` | Develop brand strategy, briefs, identity systems and brand reviews with current brand context and independent creative judgment. Use when... | | `capability-operator` | Operate Muse's capability inventory on Hatch. Routes ambiguous or multi-domain requests to the right workspace skill, inventories ~/works... | | `citizen-forge` | Turn a plain-language internal-tool idea into a governed application with deterministic risk classification, 35 named controls, and relea... | | `continuity-vault` | Keep work usable and trustworthy across sessions: route continuity operations (task handoffs, durable extraction, knowledge promotion, kn... | | `data-storytelling-studio` | Turns checked analysis into a decision-facing artifact without recalculating or overstating the source evidence. Routes analysis to the r... | | `founder-revenue-engine` | Founder-led early revenue work: turn recent public



ChatForm
Tgmlabs