Repository memory for Claude Code: index symbols and architecture, progressively retrieve context beyond the context window, and capture durable engineering decisions in a personal trove.
★not rated 1 11d agoA
tokens not measured
originalMIT
Governed, spec-driven delivery for agent-built software: operator-authorized acceptance contracts, an honestly-tiered merge floor on your own branch protection + CI, deploy-once certification, and a citation-graph MCP.
★not rated 1
changed 2d agoA
tokens not measured
originalMIT
Context-shielding, cost-tiered delegation for Claude Code: high-volume work runs in specialized subagents on the cheapest tier a grader can check, returning summaries that keep the main context small. Ships a tiered agent roster, an orchestration skill, three workflows (survey, audit, and a manifest-driven build), and.
★not rated 1 1mo agoA
tokens not measured
originalMIT
An always-on operating protocol for coding agents: build only what's asked, surface ambiguity as Q/A/P/E/X, scale process to task size, and verify before claiming. Injected into every session via a SessionStart hook.
★not rated 1 3mo agoA
tokens not measured
originalMIT
Scans your Claude Code plugin ecosystem and generates an ARCHITECTURE.md with Mermaid.js dependency graphs, collision reports, and ecosystem health scores.
★not rated 1 5mo agoA
tokens not measured
originalMIT
Eight defensive hooks: blocks recursive rm on $HOME, download-and-execute pipelines, secrets-file access, path-less formatter runs, credentials in the staged diff, and the two zsh quoting mistakes that fail silently; warns on stale lockfiles and shellcheck findings; masks credential-shaped values out of Bash output…
★not rated 1
changed 3d agoA
tokens not measured
originalApache-2.0
Routes coding labor to whichever AI CLI (codex, grok, kimi, claude/Fable) has live quota headroom and the right capability fit, runs it in an isolated git worktree, and verifies the diff before reporting.
★not rated 1
changed 2d agoA
tokens not measured
originalMIT
An explicitly invoked, token-efficient harness for planning, implementing, and reviewing agent-driven software. It helps technical and non-technical users produce sustainable code through approval-gated specifications, pragmatic repository-aware standards, conditional expert lenses, safe Git workflows, and…
★not rated 1 1mo agoA
tokens not measured
originalMIT
Cross-LLM context handoff — creates a complete handover.md so any LLM can continue your work without mistakes. Auto-triggers before context compaction.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Converts vague, messy, or under-specified requests into clearer, efficient AI-ready instructions while preserving user intent. Use before solving complex, ambiguous, or high-risk tasks.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Ask all three. Keep what works. Cross-provider councils, a taste lens, and smart routing across Claude, Codex, and Gemini — local-first, riding your existing CLI subscriptions. No new app, no service, no API key.
★not rated 1 3d agoA
tokens not measured
originalMIT