Developer Tools●●Solid
SigMap – 81.1% retrieval hit 5, 96.9% token reduce,zero deps
TF-IDF on signatures beats vector embeddings for file retrieval without the infra overhead.
Big BrainSolve My Problem
manoj079
1441mo ago
Back to browse
GitHub Repository
Token-efficient Markdown CLI for AI agents to navigate, search, and pack docs into bounded context windows
11 starsRust
Mdlens – Reduce token spend and boost retrieval on Markdown-heavy repos
by dreeseaw·Apr 26, 2026·3 points·0 comments
AI Analysis
●●SolidSolve My ProblemBig Brain
Scout command with bounded evidence packs beats naive file reads for token efficiency.
Strengths
- •Section-aware retrieval with IDs enables precise follow-up reads without full file loads
- •Built in Rust with JSON output for clean agent tool integration
- •Directly addresses token cost pain with hard max-tokens budgets per command
Weaknesses
- •Documentation retrieval is a solved category with established tools
- •Markdown-only scope limits applicability to codebase-heavy projects
Category
Target Audience
Developers building AI agents over documentation
Similar To
JinaAI Reader · Firecrawl · Context7
Similar Projects
Developer Tools●●Solid
Ledger – Claude Code Token Spend Analyzer
Local SQLite ledger captures runaway Claude Code sessions before the bill shocks you.
Solve My ProblemCozy
tsv650
301mo ago
Developer Tools●●●Banger
SkillMesh (role-based tool routing for Claude/Codex)
Context routing cuts 73% of tokens while staying 9/10 accurate on role match.
Big BrainSolve My Problem
VarunReddy023
303mo ago
Developer Tools●●●Banger
BotBrowser – MCP server, saves 90% of tokens for web-browsing agents
Strips 90% of tokens from web pages for agents—no API key, no server, MIT open source.
Solve My ProblemWizardryDark Horse
amplifyco
113mo ago
AI/ML●●Solid
TideSurf – Your agent doesn't need eyes to browse the web
Cuts agent browsing tokens by 32x with DOM-to-markdown compression and MCP support.
Big BrainShip ItNiche Gem
Mercuriusdream
102mo ago
AI/ML●Mid
Jensenify-MCP: "Every engineer should spend $250k/yr on tokens" –solved
Clever satire but injecting 2.9M tokens per call is wasteful, not useful.
Cozy
hank2000
102mo ago