fehmProduct
Code graphs, context
and change review.
Ten ways to work with the same local graph. Explore the commands and the limits of each workflow.
Understand the system
Trace files, symbols, callers, dependencies, and entry points. Find the source behind the explanation.
fehm query "authentication flow"TypeScript and JavaScript have the deepest analysis. Other supported languages have shallower resolution.
Give agents focused context
Retrieve source excerpts, related tests, architecture rules and project history within a context budget.
fehm context "add refunds" --budget 4000Token budgets are estimates for rendered context, not exact model token counts.
Check architecture and practice drift
Compare changes with approved boundaries. Detect when engineering-practice evidence disappears.
fehm practices audit31 static checks support review; they do not certify compliance or prescribe one universal architecture.
Review and verify changes
Follow a diff through dependencies, inspect its blast radius, and run supported checks.
fehm impact --base HEAD~1Verification runs project code and requires a trusted repository and its toolchain.
Find quality and reliability gaps
Connect coverage, dependency risk, API contracts, infrastructure, and runtime evidence.
fehm healthHeuristic risk reports and static test reachability are not measured runtime guarantees.
Engineer AI systems
Version prompts, review tool contracts, compare model behavior, and evaluate changes against held-out cases.
fehm ai-evaluationsLive evaluations require an explicitly configured provider. Costs and data transmission are opt-in.
Keep the team in context
Watch repositories, explore the local cockpit, preserve decisions, and export onboarding notes.
fehm serve .fehm/graph.json --watchLocal operation supports trusted teams. It is not an isolated multi-tenant service.
Testing & coverage
Inspect test quality, ingest measured coverage, and explore mutation testing and historical bug replay.
fehm test-qualityWithout a coverage artifact, test reachability is a static estimate. Test execution requires a trusted repository.
Security & delivery
Review security paths, dependency risk, API contracts, and infrastructure configuration together.
fehm security-graphSupported static patterns provide review signals, not a complete vulnerability assessment.
Review agent claims and outcomes
Check graph-backed claims and preserve agent outcomes, recurring mistakes, and resolution history.
fehm agentUnsupported claims remain unverified. Recorded outcomes do not guarantee future agent behavior.