claude

foundation

Gaia's context layer: the instruction-and-memory cascade that lets an agent understand a whole tree rather than the one file it was pointed at. Provides the whole-tree entry point that establishes, re-derives and sweeps that layer across an existing directory tree, plus single-scope context authoring, memory-store maintenance, registry reconciliation, repository-durability checks, session close-out, and the house rules for authoring new skills, agents and plans — together with the context, repo-durability and skills auditor agents. Memory, plans and lessons are files in the consuming repository — they version with the code and there is no hosted store to sync. Required by the engineering and product plugins.

33 on GitHubby FrostAura Technologiesv14.3.1MITupdated 1 month agofoundational-ai-contribution-toolset
/

Plugin foundation from marketplace frostaura in frostaura/ai.toolkit.gaia. Commands read from the repo's manifest — 1 month ago.

What it installs

Skills (10)

  • fa-foundation-context-audit

    Provides the whole-repository context-layer drift sweep — every instruction file, every MEMORY.md and topic store, and every skills directory reconciled against inspected reality rather than against what the files claim. Use it by running the mechanical audit script first, fanning out one agent per

  • fa-foundation-context-authoring

    Provides cascade-aware authoring for a scope's instruction file (CLAUDE.md or AGENTS.md) and its MEMORY.md topic store, keeping timeless rules strictly separate from dated state and holding every scope to its own delta over its parent. Use it by reading every context file in the scope end to end fir

  • fa-foundation-create-agent

    Provides the house rules for an agent definition that actually gets invoked and stays in its lane — one role per file, a description carrying what the agent is for and when to reach for it, an explicit anti-mission, an output contract, and a tool scope no wider than the role needs. Use it by reading

  • fa-foundation-create-plan

    Provides a researched, executable plan for work that is large, ambiguous or hard to reverse — requirements clarified with the human first, three identical subagents fanned out to propose plans in parallel, all three critiqued dispassionately, and the best single plan synthesized and written to a fil

  • fa-foundation-create-skill

    Provides the house rules for a SKILL.md that actually fires — naming, the frontmatter contract, a description written as a trigger rather than a summary of the body, the standard section shape, references kept one level deep, and the prior decision of whether a new skill is warranted at all. Use it

  • fa-foundation-memory-maintenance

    Provides the deep-refresh procedure for one scope's memory store — the `memory/` topic files, plus the `MEMORY.md` index regenerated from them — re-established against inspected reality instead of against what the previous version claimed. Use it by reading the whole store end to end, verifying ever

  • fa-foundation-optimize-directory-tree

    Provides the whole-tree bootstrap that puts the Gaia context layer onto an existing directory tree — an instruction file and MEMORY.md topic store at every scope that earns one, a root registry, and the retrospective design documentation its projects never had — and re-validates a bootstrapped tree

  • fa-foundation-registry-audit

    Provides a two-direction registry audit that reconciles a repository's registry of scopes against the directories that actually exist on disk. Use it by running the context-audit script for the mechanical forward and reverse passes, then cross-checking by hand every other registry that names the sam

  • fa-foundation-repo-durability

    Provides the six-probe git inspection that establishes whether a repository is genuinely durable and genuinely safe to push, by running commands rather than trusting what a context file claims. Use it by running each probe in order, classifying the upstream topology in words, reading what the unpush

  • fa-foundation-session-close

    Provides the end-of-session close-out that brings a scope's instruction file, MEMORY.md topic store and local skills back into line with what actually changed, and empties the session's scratch directory. Use it by walking every touched scope in order — memory store, then rules, then skills, then re

Subagents (3)

  • fa-foundation-context-auditor

    Use for reconciling ONE scope's context layer — its instruction file, its memory store and any local skill definitions — against what is actually on disk, rather than what those files claim. This role owns the corrections inside its assigned scope: rewriting stale memory topics, regenerating the der

  • fa-foundation-repo-durability-auditor

    Use for establishing whether a repository is genuinely durable and genuinely safe to push, by inspecting git rather than trusting what a context file, a README or an earlier report claims about it. This role owns the durability evidence — branch and upstream, ahead/behind, divergence, unpushed and d

  • fa-foundation-skills-auditor

    Use for auditing one skills-and-agents directory — its `SKILL.md` files, its agent definitions and any scripts beside them — for staleness, sprawl, duplication, dead pointers and descriptions that never fire. This role owns that directory outright: it verifies, rewrites, merges, deletes and re-index

MCP servers (1)

  • fa-gaia-remote

    connects to: https://gaia.frostaura.net/mcp

View source on GitHub gaia.frostaura.net

More from frostaura/ai.toolkit.gaia

All 3 plugins →

product

claude

A money-gated lifecycle for taking consumer one-off / in-app-purchase products from idea to validated bet through launch and live-ops. Provides 10 stage playbooks (discovery, ideation, monetization design, validation, vertical-slice/MVP, production hardening, compliance, soft-launch, GTM launch, live-ops, and portfolio retrospective) plus a shared money-gate and unit-economics discipline, orchestrated by a coordinator and specialist, reviewer, and synthesizer agents that fan out, adversarially review, and synthesize each stage against net-of-fee economics.

14 skills · 11 agents

33

Gaia's spec-driven software delivery lifecycle: intake, architecture, planning, implementation, testing, and release. Provides stage playbooks (architecture documentation, default tech stack, execution planning, engineering delivery, UI engineering, formal testing, and end-to-end process orchestration) plus the specialist agents (intake coordinator, solutions architect, implementation planner, software engineer, tester, release engineer) that take a goal from request to validated, released software through enforced QA and release gates.

16 skills · 6 agents · 1 MCP server

33

Similar plugins

AI-powered codebase understanding — analyze, visualize, and explain any project

9 skills · 10 agents · hooks

Egonex-AI/Understand-Anything 92,410

career-ops

claudecodex

Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications

5 skills · hooks

career-ops-hq/career-ops 82,253

humanizer

claude

Rewrite AI-sounding text so it reads like the writer without changing what it says.

1 skill

blader/humanizer 68,461

mempalace

claudecodex

AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.

3 skills · 6 commands · 1 MCP server · hooks

MemPalace/mempalace 60,536

Perform AI code review on Git diffs — supports workspace changes, branch ranges, and single commits with concurrent per-file analysis, codebase search, and deep context-aware review.

2 commands

alibaba/open-code-review 50,068

llm-finetuning

claudecodex

Eval-gated LLM fine-tuning lifecycle (LoRA/QLoRA, DPO, GRPO/RLVR, vision SFT, quantized export). Use when fine-tuning an open-weights model end-to-end: building the eval harness first, selecting method and model, preparing data, training with Unsloth or TRL, gating checkpoints, and exporting GGUF/FP8.

10 skills · 2 commands · 3 agents

wshobson/agents 42,461

Maintain foundation?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins3 plugins