By acmelabs-15
ACMElabs Brain skills: composition library (SHA-256 round-trip decompose/recompose) + curator skills (defrag/ingest) + Brain lifecycle skills (plan/research/decisions/spec/build/review/end)
This skill should be used when the user asks to "build X", "implement the SPEC", "code up SPEC-NNN", "build out the implementation", "do the build phase", "start coding", "write the code for X", "implement the tasks", "execute the build", "start building", or invokes /build in any form. Also auto-invoked by /plan PLAN-NNN when the next-ready part is a build.SPEC-NNN phase. Runs the per-TASK atomic cycle (Stage A: implementer dispatch with TDD plus canonical-source-mirror plus evidence-hierarchy briefs, then qa dispatch, then state propagation, then sync-jira push, then PLAN tick plus commit), then a final spec-level QA sweep with coverage matrix (Stage B), then EPIC plus sync-jira plus PLAN propagation (Stage C), then 4 mandatory exit gates (code-qualities-assessment, incoherence, orphan-ref coverage-note, lint) plus conditional Gate 5 prompt-engineer for any diff containing prompt-like content, before flipping SPEC ACCEPTED then IN_PROGRESS then DONE.
This skill should be used when the user asks to "lock the decisions", "adjudicate the analysis", "make decisions on X", "decide on the options", "author the ADR for X", "go through the decisions phase", "pick from the options", "what should we pick for X", "review the analysis and decide", "finalize the architecture decisions", or invokes /decisions in any form. Also auto-invoked by /plan PLAN-NNN when the next-ready part is a decisions.N phase. Adjudicates options surfaced during /research — runs the per-D-N micro-cycle (decision-critic stress-test, AskUserQuestion with verbatim options from ANALYSIS notes, verbatim echo, diff-approval, 2-step PLAN-then-SESSION edit, commit), then authors a composite ADR via architect dispatch with detail-parity mandate, then runs the adr-review skill as MANDATORY blocking gate before flipping ADR PROPOSED to ACCEPTED.
Split (1-to-N) a Brain knowledge-graph note into multiple notes using a distribution plan. Triggers on phrases like "decompose this note", "split this ADR into per-decision notes", "break this analysis into clusters". Authors a distribution plan YAML, presents it for adjudication, then executes via the composition library with SHA-256 round-trip validation guaranteeing zero content drift.
Periodic curator skill for a Brain knowledge graph. Audits notes under docs/** against quality thresholds and scope-evaluation heuristics, classifies them into split / merge / stale / structural-fix candidates, then delegates restructuring to the decompose and recompose primitive skills (and to native delete for stale entries). Supports interactive mode (user confirms each candidate) and report-only mode for cron scheduling.
This skill should be used when the user asks to "end the session", "wrap up", "wrap up the session", "ship the changes", "create the PR", "close out", "finish the session", "call it done", "open the PR", "submit the work", or invokes /end in any form. Runs the session-end pipeline: verify all part DoD checkboxes are ticked, invoke /review as the BLOCKING gate, on review FAIL surface 3-option AskUserQuestion (fix-and-retry, defer-with-rationale, abandon-session), on review PASS run 5 pre-flight checks (secret-scan, tests passing, lint clean, PR description validation, CI workflow health), then commit final state, markdown lint fix, push branch, create PR via gh pr create, flip session status DONE, and emit a structured end-of-session report with pre-flight results plus review verdict plus PR link plus warnings plus next-session recommendation.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Zero-content-drift restructuring + ingest skills for Brain knowledge graphs.
This project provides four Claude Code skills:
| Skill | Role | Primitive / Composer |
|---|---|---|
/decompose | 1 → N split with SHA-256 hash-validated content preservation | primitive |
/recompose | N → 1 merge with SHA-256 hash-validated content preservation | primitive |
/defrag | Periodic curator; audits the graph and delegates to /decompose, /recompose, delete | composer |
/ingest | Bring external content into the graph as a CONVENTIONS-compliant note | composer |
The decompose and recompose skills are documented in their own SKILL.md
files; this README focuses on the two new composer skills, /defrag and
/ingest, plus the shared install procedure.
./install.sh
The script creates symlinks at ~/.claude/skills/ for the four skills owned by
this project (/decompose, /recompose, /defrag, /ingest). It is
idempotent for its own symlinks (ln -sf) and never touches symlinks under
~/Dev/basic-memory-skills/.
/defrag audits the active project's notes under docs/** against quality
thresholds and scope heuristics, classifies each note into split / merge /
stale / structural-fix candidates, then delegates the restructuring work to
/decompose, /recompose, or native Brain MCP delete.
| Flag | Effect |
|---|---|
| (none) | Interactive mode: print report, confirm each candidate |
--report-only | Cron mode: write the report to defrag/reports/defrag-YYYY-MM-DD.md and exit |
--project-root <dir> | Set the project root (default: cwd) |
--staleness <days> | Staleness threshold in days (default: 90) |
--basic-memory | Treat the project as basic-memory; skip CONVENTIONS-specific checks |
docs/**.> 15 observations without H3 sub-grouping → split candidate> 500 lines with multi-entity content → split candidate< 3 observations OR < 2 relations → merge candidate> 12 relations without H3 type-grouping → structural-fixDONE /
DEPRECATED → stale candidate# Interactive
bun skills/defrag/scripts/defrag.ts
# Cron / report-only
bun skills/defrag/scripts/defrag.ts --report-only --staleness 90
# Specific project
bun skills/defrag/scripts/defrag.ts --project-root ~/Dev/brain --report-only
/ingest parses an external markdown file (or directory of files), detects
the appropriate canonical entity type, and writes a well-formed Brain note via
the Pattern 2 three-phase write. The source body is preserved character-for-
character between the frontmatter block and the generated Observations
section.
| Flag | Effect |
|---|---|
<path> | Source file (or directory with --batch) |
--type <type> | Force entity type; one of the 16 canonical types |
--parent-spec <id> | Required for spec-nested types (requirement, design, task) |
--basic-memory | Simplified write; skip CONVENTIONS-specific requirements |
--batch | Treat <path> as a directory; ingest each .md file |
--dry-run | Show the plan; do not write |
--project-root <dir> | Project root (default: cwd) |
--descriptor <text> | Override the descriptor used in the title |
## Observations / ## Relations sections.type → --type flag → fallback
(analysis).## Observations (≥ 3 with category prefix
## Relations (≥ 2 with valid types) when absent.write_note with a
space-separated title (no colon); Phase 2 edit_note find_replace to add
the colon in frontmatter title and H1; Phase 3 move_note to rename to
the kebab filename.npx claudepluginhub acmelabs-15/skills --plugin skillsUI/UX design intelligence. 67 styles, 161 palettes, 57 font pairings, 25 charts, 15 stacks (React, Next.js, Vue, Svelte, Astro, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Nuxt, Jetpack Compose). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
This skill should be used when users need to generate ideas, explore creative solutions, or systematically brainstorm approaches to problems. Use when users request help with ideation, content planning, product features, marketing campaigns, strategic planning, creative writing, or any task requiring structured idea generation. The skill provides 30+ research-validated prompt patterns across 14 categories with exact templates, success metrics, and domain-specific applications.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.