Plugin Claude Code
Bundles 8 skills, 2 commands · 296 tokens together
Spec-driven development methodology for Claude Code. Provides skills for requirements engineering (EARS format), design documentation, task breakdown, AI prompting strategies, quality assurance, and troubleshooting.
Skill Claude CodeCodex
Search across rotated log archives ($LAZARHOME/logs/stream.jsonl..bak) when you can't find what you need in current stream.jsonl or in memory/log-summaries/.
Skill Claude CodeCodex
When you need a capability that isn't in INDEX.md, author a new skill.
Skill Claude CodeCodex
Scaffold a terminal interface for lazar — a small client project under workspace/ / that consumes lazar -p --output-format stream-json and renders the events. Default stack: TypeScript + readline. Optional: Go + Bubble Tea, Python + Textual, or anything else the user wants. Use when the user asks for a "TUI", "chat…
Skill Claude CodeCodex
Use lazar recursion to extract durable, structured learnings from rotated log archives, and write them into categorized memory files. Turns raw events (L3) into curated knowledge (a richer L2) on demand.
Skill Claude CodeCodex
Load, list, verify, and store API keys. Canonical source is macOS Keychain (service=lazar). Use when you need a credential, to check which providers work, or after rotating keys. Never print secret values.
Skill Claude CodeCodex
Decide whether a capability already exists, can be composed from existing skills, or needs a new skill.
Skill Claude CodeCodex
Continual harness — ranked scoped store for memories, skills, prompt notes, and subagent recipes. Auto-injected at session-start. Use to record durable lessons (add), load full entries, score helpful/unhelpful, and seed from playbook/learnings. Prefer this over freeform notes.md appends for anything that should affect…
Skill Claude CodeCodex
Verify lazar is installed correctly and all load-bearing behaviors work end-to-end. Runs static checks (kernel locked, env vars, skills present), boundary tests (sandbox enforcement), live behavior tests (log auto-rotation, memory persistence, recursion), and short end-to-end builds in TypeScript and Python. Runs in a…
Skill Claude CodeCodex
Recover relevant prior conversation without blowing your context window. Lazar has three tiers of memory — work through them in order, only going deeper when shallower tiers don't have the answer.
Skill Claude CodeCodex
Turn raw stream.jsonl logs into durable, readable memory — a work journal of what I built and worked on — so the multi-MB archives stop being the only record of my own history. Run after sessions, before pruning logs, or whenever I can't recall recent work. Pairs with meta/distill (the learnings layer). Use when logs…
Skill Claude CodeCodex
Add a richer memory summary for a rotated Lazar stream archive.
Skill Claude CodeCodex
Before modifying any code in a project directory, read the project's PROJECT.md (if present) to learn its conventions, structure, and rules. This prevents the most common drift: the agent inventing a parallel implementation instead of editing the one that exists.
Skill Claude CodeCodex
When you identify a change that needs to happen in the immutable kernel (bin/lazar or src/), you cannot apply it yourself — sandbox-exec blocks writes there, and chflags uchg blocks them at the OS level too. Instead, stage a complete proposal in workspace/ that the user can review and apply with one command.
Skill Claude CodeCodex
How I notice a recurring failure of my own, root-cause it, fix what I can myself (behavioral rules, config, ignore files), and — for anything needing a kernel change I can't apply — draft a verified proposal for the operator to review. The capstone that turns my mistakes into durable fixes. Use when I hit a repeated…
Skill Claude CodeCodex
Make a skill or wrapper invokable from the user's shell as a command, without writing to PATH directories. The sandbox blocks writes to /bin/, /usr/local/bin/, etc. by design — that boundary exists so the agent never silently defines what happens when the user types a command. Hand off through an explicit aliases file…
Skill Claude CodeCodex
The ONLY sanctioned way to delegate work to a lazar sub-agent. Uses locked bin/lazar-spawn — lints the prompt, bounds retries, wall-clock-kills wedged children, and gates success by re-checking declared claims against the filesystem. Prefer pack-and-spawn.sh to write a BRIEF context pack then dispatch. Raw lazar -p is…
Skill Claude CodeCodex
Turn lazar's enforcement layers on or off. Use when a write fails with 'Operation not permitted', when the operator asks for full access, when you need to edit bin/ src/ logs/, or when the operator asks to re-enable safety. Toggles all four layers at once: sandbox-exec, kernel binary lock, src/ read-only, and pre-tool…
Skill Claude CodeCodex
Durable notes that survive across lazar -p invocations. Memory lives in $LAZARHOME/memory/ and is plain markdown — no JSON, no encoded blobs. You and the user should be able to cat it and understand instantly.
Plugin Claude Code
Bundles 1 skill · 0 tokens together
Collection of Claude Code skills, MCP servers, and hooks for enhanced productivity.
Plugin Claude Code
Bundles 1 skill, 3 commands · 24 tokens together
Learn from conversation history to improve Claude Code performance. Analyzes logs for communication patterns, model understanding, assumptions, efficiency, and extracts actionable lessons.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: