Library
85,762 resources from 11 public sources. 53 checked by hand; the rest are unreviewed.
Results
-
Prompt-cache prefix guardian
MCP serversCache-bust forensics + TTL-fit verdicts: names the field that busts your prompt cache.
-
killi
MCP serversOpen-source AI coding agent for your terminal, written in Rust. One 7 MB native binary for Apple Silicon Macs, ~18 MB RAM, 70–90% prompt-cache hits. Works with OpenAI, Anthropic and any compatible LLM; MCP, subagents, plan mode, skills, hooks, AGENTS.md. A lightweight Claude Code / Codex CLI alternative.
git clone https://github.com/timwuhaotian/killi.git -
TokenSafe
MCP serversReduce your AI bills by up to 99% locally: focused code, warm prompt caches & zero secret leaks.
tokensafe@4.1.6 -
token-optimizer-mcp
MCP servers91.5% fewer tokens on tool output on Headroom's own benchmark, where Headroom removes 61.4%. Separately, the same SWE-bench tasks resolved at 0.71x the cost of plain Claude Code. Plus memory across sessions and subscription quota pacing. Claude Code, Codex, Copilot, Gemini and 12 more. Local, MIT.
git clone https://github.com/ooples/token-optimizer-mcp.git -
DoboCache
MCP serversStabilizes LLM prompt-cache prefixes so provider caching hits. One command setup, no baseURL edit.
dobocache@0.9.2 -
Claude API Pricing
MCP serversCurrent Claude API prices and cost estimates with prompt caching and Batch API pricing.
-
Etuki FinOps & LLM Pricing Oracle
MCP serversCryptographic 563-leaf LLM rate oracle, context-cliff auditor, and prompt cache optimizer.
-
OmniToolCalc LLM Token & Cloud FinOps Suite
MCP serversEstimate 2026 enterprise LLM API token costs, prompt caching savings, and developer benchmarks.
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.