By atuljha23
A Jedi council for your codebase: 14 role-specialist agents, 24 domain commands (a11y, perf, QA, security, threat-modeling, SRE, ADR, cost tracking, MCP auditing), 11 skills including token economy, and a SQLite-backed learning crystal that compounds your corrections across sessions.
Accessibility audit (WCAG 2.2 AA) on the working-tree diff or a specified file/directory. Delegates to @a11y-auditor.
Draft an Architectural Decision Record (ADR) for a proposed or accepted decision. Uses the MADR-style template and places it in docs/adr/ or adrs/ (whichever the repo uses).
Set a budget and get nudged when you're approaching it.
Do these in order. Do not skip.
Snapshot the current session's context footprint — transcript size, CLAUDE.md size, MCP schemas, file reads still in context. Identifies what's eating tokens so you can trim.
Accessibility audit against WCAG 2.2 AA. Use when the user asks "is this accessible", is building UI components, or when /holocron:a11y runs. Flags missing semantics, ARIA misuse, focus order, keyboard traps, and likely contrast issues.
System design, tradeoff analysis, and ADR drafting. Use when the user needs to pick between implementation approaches, evaluate a design, decompose a feature into modules, or document a decision for posterity.
Implements and reviews server-side code — HTTP APIs, services, background workers, data access. Use for endpoint design, request/response shape, error handling, database queries, async/concurrency patterns, and service boundaries.
Staff-level PR review of the current diff or a specified ref. Use when the user asks for "a review", "a second opinion", or runs /holocron:review. Reviews for correctness, clarity, test quality, and blast radius.
Analyze session-usage logs and identify concrete token-savings opportunities. Reads ~/.holocron/sessions/*.json, correlates cost to behavior, and proposes changes ranked by expected savings. Use when costs creep up or before setting a budget.
WCAG 2.2 AA quick reference — semantics, keyboard, names, states, contrast, motion, errors. Use when building or reviewing UI components.
REST and GraphQL API design conventions — resource modeling, pagination, error envelopes, versioning, idempotency. Use when adding, changing, or reviewing endpoints.
Apply the Write/Select/Compress/Isolate framework to manage the context window. Use when the task is long, the transcript is bloated, the model is drifting, or before /holocron:handoff.
Hypothesis-driven debugging — reproduce, hypothesize, probe, locate, fix, guard. Use when something is broken and you don't yet know why.
Safe schema-migration patterns for systems under live traffic — expand/contract, backfill, double-write, shadow-read, online DDL. Use when authoring, reviewing, or sequencing a migration that can't take a maintenance window.
Matches all tools
Hooks run on every tool call, not just specific ones
Executes bash commands
Hook triggers when Bash tool is used
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.
Modifies files
Hook triggers on file write and edit operations
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
A Claude Code plugin that gives you a Jedi council of role-specialist agents, domain commands (a11y, perf, QA, security, ADR), and a SQLite-backed learning crystal that compounds your corrections across sessions.
A holocron, in Star Wars lore, is a crystalline cube that stores Jedi teachings. This one stores yours — every correction you make becomes durable knowledge the assistant carries into the next session.
atuljha23.github.io/holocron · v0.3.0 · MIT · targets SWEs, frontend engineers, backend engineers, QAs, SREs, and security.
If you use Claude Code daily you already know the failure mode: you correct the assistant the same way a dozen times, every session feels like starting over, and the "plan → review" discipline rots the moment compaction hits. Holocron does three things about that:
@frontend-engineer, @qa-engineer, @a11y-auditor, @security-reviewer, @perf-engineer, …) each with a sharpened rubric and clear handoff boundaries.# Add this repo as a marketplace and install the plugin
/plugin marketplace add atuljha23/holocron
/plugin install holocron@holocron
# One-time: install the native dependency the learning crystal needs
cd ~/.claude/plugins/marketplaces/holocron && npm install
Local dev against a clone:
git clone https://github.com/atuljha23/holocron ~/projects/holocron
cd ~/projects/holocron && npm install
/plugin marketplace add ~/projects/holocron
/plugin install holocron@holocron
Health check:
/holocron:doctor
| Component | Count | Where |
|---|---|---|
| Role / meta agents | 14 | agents/ |
| Slash commands | 24 | commands/ |
| Skills | 11 | skills/ |
| Hook scripts | 13 | scripts/ |
| Hook events wired | 6 | hooks/hooks.json |
| Context profiles | 4 | contexts/ |
| Rules | 4 | rules/ |
Run node scripts/holocron-cli.js inventory anytime to verify counts match this table.
v0.3.0 — Token Economy. Every Claude Code session's $ and tokens are logged to
~/.holocron/sessions/. Run/holocron:costfor the rolling total,/holocron:budgetto set caps,/holocron:mcp-auditto cut schema bloat,@cost-analystfor ranked savings proposals.
/holocron:onboard # draft a CLAUDE.md for this repo
/holocron:plan # enter plan mode with a real scaffold
/holocron:develop <feature> # research → plan → implement → review
/holocron:commit # quality-gated commit with a reviewed diff
/holocron:learn "<rule>" # capture a correction so it sticks
@frontend-engineer <task> # role-specialist for component/UX work
/holocron:a11y # audit changed UI files against WCAG 2.2 AA
/holocron:perf # bundle / LCP / CLS check on the diff
/holocron:test-gap <file> # find the component states nobody tests
@backend-engineer <task> # API / data / service work
/holocron:sec-scan # SAST + secrets + dep advisories on the diff
/holocron:perf # N+1 and index-gap sweep
/holocron:adr "<title>" # record a real decision, not a Slack message
@qa-engineer <task> # test strategy / coverage / flakes
/holocron:test-gap <file> # prioritized list of missing tests
/holocron:flaky <test> # walk the flake-triage tree with evidence
@sre-engineer <task> # Docker / k8s / CI / runbooks
/holocron:llm-gate "push to main"
/holocron:permission-tune # reduce prompt fatigue with allow/deny tuning
@threat-modeler <feature> # STRIDE on a proposed change
/holocron:threat-model # design-level review
/holocron:sec-scan # SAST + secrets + CVE on the diff
/holocron:cost --breakdown # $ + tokens + cache hit rate
/holocron:budget --session 2 --daily 10
/holocron:mcp-audit # MCP token overhead
@cost-analyst # ranked savings from usage data
All live under agents/ and invoke as @<name>:
npx claudepluginhub atuljha23/holocron --plugin holocronHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
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.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
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.