donaldfilimon/abi / .agents/skills/abi-doc-claims-sync
abi-doc-claims-sync
VerifiedSynchronize ABI Markdown and docs claims with executable behavior. Use before editing README, AGENTS.md, CLAUDE.md, GEMINI.md, docs/**/*.mdx, walkthroughs, changelogs, or any old ABI Markdown that mentions WDBX, MCP, GPU, clustering, compression, FHE, agents, or benchmarks.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 1.6 KB |
| references/claim-boundaries.md | markdown | 3.3 KB |
references/claim-boundaries.md
View on GitHub# ABI Documentation Claim Boundaries
Authoritative policy files:
- `docs/contracts/external-claims-audit.mdx`
- `docs/contracts/public-api.mdx`
- `docs/spec/wdbx-north-star.mdx`
- `abi-threat-model.md`
Safe repo-backed wording:
- ABI is a Rust 2024 local AI orchestration framework targeting stable 1.99.1 with deterministic Abbey/Aviva/Abi profile routing.
- WDBX is an in-process vector/key-value/block store with segment checkpoints, WAL recovery, temporal/causal records, HNSW-style cosine search, and hybrid ranking where wired.
- CLI and MCP surfaces are frozen and contract-tested.
- MCP stdio is local IPC. Startup also attempts a custom loopback HTTP listener
that can require `ABI_MCP_HTTP_TOKEN`. It serves persistent MCP 2024-11-05
HTTP+SSE sessions (contract-tested against stdio bytes) plus a one-shot
direct-reply `POST /message` mode; it is not Streamable HTTP (2025-03-26).
- WDBX REST is loopback and can require `ABI_WDBX_REST_TOKEN`.
- WDBX cluster RPC uses real TCP RequestVote/AppendEntries, supports `ABI_WDBX_CLUSTER_TOKEN`, optional `ABI_WDBX_CLUSTER_PEERS`, and refuses non-loopback binds without a token. Signed membership, rendezvous placement, exact committed-transaction replication/read repair, resumable rebalance, and `cluster local-demo` are locally tested; they are not production separate-host proof.
- GPU support includes capability/status reporting and vector operations with
linked Metal fused cosine/dot/L2 plus multi-pass `reduce_sum_kernel` on macOS
when native kernels initialize, else deterministic CPU fallback.
- Versioned PQ and persisted-autoencoder artifacts are deterministic, validated, locally tested codecs; they are not SOTA or production-quality claims.
- DGHV refresh is secret-key-assisted and educational, not cryptographic bootstrapping. The optional pinned TFHE-rs demo executes boolean/integer/programmable-bootstrap APIs but has no ABI-independent cryptographic audit or protective-use approval.
Claims to remove, downgrade, or frame as proposed:
- Distributed sharding or production multi-host database deployment.
- AES/RBAC WDBX storage unless implemented and tested.
- Regulatory certifications.
- Kubernetes/H100/A100/InfiniBand/NVLink deployment claims.
- QPS, latency, accuracy, speedup, energy, empathy, SQuAD, or CodeSearchNet numbers without fresh artifacts.
- Native local CUDA/Vulkan/ANE execution when the source only proves detection, reporting, or fallback (macOS Metal map + multi-pass reduce is real when initialized — still not a blanket GPU speedup claim).
- Production/SOTA learned compression.
- ABI-authored production-secure FHE, independent cryptographic audit, or protective-use approval. A passing optional TFHE-rs demo is narrower evidence.
- Production-ready non-loopback MCP/WDBX HTTP without TLS, authz, rate limits, and deployment controls.
Instruction-file sync checklist:
- Command list and forbidden legacy names.
- WDBX grammar including `db compact`, `cluster local-demo [nodes] [--json]`, `cluster serve <port> [node] [host]`, `secure dghv-bootstrap|tfhe`, and `api serve [port]`.
- MCP 12-tool names and transport details.
- Feature flags and FoundationModels gating.
- Stable-Rust/toolchain-neutral idioms and `./tools/cargo.sh` vs bare `cargo` caveat.
- `./tools/check.sh` (primary gate), `./build.sh check` compatibility alias, docs validation, parity, and cross-smoke guidance.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |