Give your agent the right context — up to ~97% fewer tokens
SigMap turns a whole repository into verified function & class signatures — the map an AI coding agent actually needs. No more pasting files or blowing the context window. Try it on any public repo below.
NewMeasured across 405 repos: ~99% fewer tokens, 96× cheaper context — reproducible numbers, with the honest agent results too.See the proof →- 97%
- fewer tokens
- 75.6%
- hit@5 retrieval
- 52.2%
- more tasks solved
- 0/21
- context overflows (was 16/21)
SigMap v7.25 benchmark · 21 repos · 90 retrieval tasks · sigmap.io
Paste a repo
Any public GitHub repository. SigMap detects the real source folders for you.
SigMap extracts signatures
Function & class signatures only — secrets redacted, ranked by relevance, full coverage.
Feed your agent
Up to ~97% fewer tokens. Copy the context file, ask the codebase, or run npx sigmap in your repo.
Everything SigMap does
full docs ↗Signature map
try abovenpx sigmapTurns a whole repo into verified function & class signatures — ~97% fewer tokens.
Ranked retrieval
try abovesigmap askAsk in plain English, get the exact files that matter — TF-IDF ranked, no LLM needed.
Coverage validation
try abovesigmap validateConfirms the right files are in scope before you trust the context.
Groundedness judge
in CLIsigmap judgeScores whether an AI's answer is actually grounded in your code — the verification layer.
Local learning
in CLIsigmap learnReinforces the files that proved helpful, so retrieval gets sharper over time.
Secret redaction
try abovebuilt-inScans signatures for AWS keys, tokens, DB strings and redacts them automatically.
33+ languages
try aboveautoTypeScript, Python, Go, Rust, Java, Kotlin, Ruby, PHP, Swift, C#, C++, Vue, Svelte…
7 agent adapters
try above--adapterEmits for Claude Code, Cursor, Windsurf, Copilot, Codex, OpenCode and Gemini CLI.
MCP server
in CLIsigmap --mcpModel Context Protocol ready — agents pull ranked context on demand.