Use fewer agent tokens
Give coding agents a compact, reusable model of the repository instead of making them rediscover structure and history on every task.
Architecture-aware context · CLI and MCP access
See agent contextOpen source · Source-verified · Benchmarked
Codebase intelligence that shows its work.
Give AI agents reusable codebase context that cuts token use. Find code-health and change-risk hotspots before you edit, and automatically generate architecture documentation that stays current.
Browse public repositories without an account. Sign in for 10 grounded answers each month.
Or explore react/reactBenchmark leader
Retrieval + agent efficiency
0.876#1 file coverage
ContextBench, 42 sealed instances, Python and Go
31.6%#1 agent-loop reduction
django/django at a single pinned commit
0.74defect-prediction ROC AUC
21 repositories, 9 languages, 2,826 files
One shared codebase index
One index. Four practical outcomes.
Repowise connects source structure, git history, health signals, tests, and engineering decisions once. The same evidence then powers agent context, documentation, change risk, and code health.
Generate documentation that stays current
Build documentation from symbols, modules, dependencies, and decisions, then identify when the underlying source has changed.
Architecture and module pages · Source-linked explanations
Explore documentationSee change risk before the diff
Trace dependents, historical co-changes, ownership, and missing tests before a developer or agent edits a high-impact file.
Blast-radius analysis · Test and co-change gaps
See change riskFind where code health is declining
Rank defect risk, maintainability problems, and static performance findings, then move from the score to the exact file and next step.
Defect-validated scoring · File-level findings
See code healthFrom index to everyday work
See what Repowise gives your team after the index is built.
For coding agents
Give agents precise context without resending the repository.
Index the codebase once, then let MCP retrieve the exact symbols, callers, decisions, health signals, and documentation an agent needs for the task in front of it. The agent spends less of its context window rediscovering the repository.
In our published Codex agent-loop benchmark, Repowise cut output tokens by 31.6%.
n = 43 · p = <0.0001 · Codex (gpt-5.6-sol) · django/django at a single pinned commit · 2026-08
Read the agent benchmark
For safer changes
Find risky, unhealthy code before it slows a change.
See defect risk, maintainability, performance, churn, blast radius, and open findings in one evidence model. Drill from the repository score into the files and dependencies that make a change risky.
Scores are computed from repository evidence and checked against real bug-fix history—not generated as an untraceable opinion.
Explore code health
For shared understanding
Generate documentation that stays tied to the source.
Turn the index into navigable documentation for the repository, its modules, and its files. Freshness status and source links make it clear what the page describes and when it needs to be regenerated.
Use the same documentation for onboarding, architecture reviews, and agent retrieval instead of maintaining separate snapshots.
Explore generated docs
GitHub PR bot
Bring codebase context into every pull request.
Install the GitHub App to review a diff against the indexed codebase, its dependency graph, and its change history—where your team already decides what ships.
- Changed contracts
- See callers outside the diff before an interface change merges.
- Missing co-changes
- Catch files that history says usually move with this change.
- Merge checks
- Put deterministic repository evidence directly in the PR workflow.
Run it your way
One engine. Three ways to operate it.
Start locally, use the hosted product, or bring Repowise into an organizational environment without changing the underlying codebase model.
Open source
Index code locally, connect your tools over MCP, and keep control of the runtime and model provider.
View on GitHubHosted
Browse public repositories for free, sign in for 10 grounded answers each month, and add private repositories when you need them.
Start freeEnterprise
Operate Repowise for an organization with the controls, support, and deployment model your environment requires.
Explore enterpriseFREQUENTLY ASKED
Questions, answered
What does Repowise index?
Repowise analyzes source structure, dependencies, call relationships, git history, ownership, code-health signals, tests, and engineering decisions. It turns that evidence into a reusable codebase model and generated documentation.
Can I browse without signing in?
Yes. You can open and browse supported public repositories without an account or installation.
What do I get when I sign in?
The Free hosted experience includes 10 grounded answers each month. Those answers use indexed repository evidence rather than treating the codebase as an unstructured prompt.
Does Repowise replace my coding agent or IDE?
No. Repowise supplies codebase context to developers and existing tools through the hosted application, CLI, and Model Context Protocol. Your agent or editor remains where the work happens.
Can I run it myself or inside my organization?
Yes. The core engine is open source under AGPL-3.0 and can run locally with your model provider or a local model. Repowise is also available as a hosted product, with an enterprise path for organizational deployments.
Start with a real repository
Start with the code. Get to the answer faster.
Explore a public repository now, or sign in to ask 10 grounded questions each month.