A set of working rules for Claude Code, Anthropic’s coding assistant, aimed at keeping long-running development work on track. It checks destructive actions, grounds work in evidence, and avoids claims that cannot be supported.
Bootstrap a project's agent documentation system in one shot: deploys the CLAUDE.md + .claude/rules/ + .agent-workspace/ trio plus the skill-authoring toolkit. Built on a neural-network doc model — context economy, growth-from-evidence, and a genome/phenotype heredity mechanism.
★not rated 1
changed 2d agoA
tokens not measured
originalMIT
Portable agentic-coding extensions — slash commands, subagents, and skills for Claude Code, including /optimise to audit a project's Claude Code, Claude Design, or Codex setup against a lessons library. Hooks and the transcript→lessons pipeline live in the full repo (see README); this plugin is the lightweight…
★not rated 1 8d agoA
tokens not measured
originalMIT
Claude Code injects your account email into session context, where an agent may reuse it as a legitimate value and commit it. This warns once per session when a personal address appears in a tool payload. Never blocks.
★not rated 1 1mo agoA
tokens not measured
originalMIT
Optional companion to warden. Re-injects your hard rules on every prompt so they survive context compaction. Adds a UserPromptSubmit hook that reads one local file and writes nothing.
★not rated 1 1mo agoA
tokens not measured
originalMIT
Branded presentation deck toolkit. Ships three skills: deck-brand-init (brand.yaml scaffolding), deck-create (conversational HTML deck authoring), and deck-publish (HTML to PDF or PPTX via Puppeteer + pptxgenjs, both pixel-perfect screenshot and fully-editable modes). Works in Claude Code and GitHub Copilot CLI.
Auto-detects API keys and secrets in your prompts, saves them to chmod-600 files in /.claude/secrets/, blocks the original prompt, and re-submits a sanitized version via OS-level paste automation. Cross-platform (macOS, Linux, Windows). Zero-friction secret hygiene for Claude Code.
★not rated 1 27d agoA
tokens not measured
originalMIT
End-of-Turn Queue — queue prompts with /queue:add and each is delivered back to Claude at the end of the turn, after it fully finishes its work, never mid-task. A FIFO prompt queue for Claude Code, built on a self-terminating Stop hook.
★not rated 1 3mo agoA
tokens not measured
originalMIT
Session continuity across context resets for Claude Code. Snapshots a session digest before /clear, /compact, or auto-compaction and auto-rehydrates it after. The non-looped companion to cc-repete — stands down whenever a cc-repete loop is active.
★not rated 1
changed todayA
tokens not measured
originalMIT
Checkpoint workflow for long-running projects: at clean breakpoints, commit, save durable memories, refresh the living design doc (source-verified), and emit a post-compaction rehydrate go-block.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Create new Claude Code skills or slash-commands in a clean, consistent house style — XML phases, state files, saved rules, checkpoint resumption, and 14 compaction patterns. Structured intake -> generate -> validate.
★not rated 1 1mo agoA
tokens not measured
originalMIT
The Fable Method: systematic unknowns-discovery workflow based on Thariq's 'A Field Guide to Fable'. Blindspot scanning, structured interviews, prototype-first, implementation notes with deviation tracking, and merge quiz gates.
★not rated 1 1mo agoA
tokens not measured
originalMIT
Advisor Strategy skill for Claude Code — spawns a stronger model (Opus) as a strategic advisor for architecture, security, debugging, and performance decisions.
Finds screenshots that pile up during each Claude Code session and, before you leave, asks to move them to the Trash. macOS + Linux. Per-session scoping, safe (recoverable) deletion, zero dependencies.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Structured planning for Claude Code with interactive decision gates. Walks through requirements gathering, approach selection, and plan validation before any code is written. Two modes: task-based (single plan file with sequential phases) for straightforward features, and spec-driven (multi-file specs with dependency…
★not rated 1 6mo agoA
tokens not measured
originalMIT
Skills that pair with the Ranql MCP to run a law firm's marketing end to end — competitor ad intelligence, brand identity, organic content analysis, and Canva creative generation.
Lovable sound cues for your coding agent — a different voice each session tells you when it starts, finishes, errors, or needs you.
★not rated 1 2mo agoA
tokens not measured
originalMIT
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: