Getting Started: Landmark for Engineers

Landmark shows you your own practice evidence and growth data. It compares your engineering record against the markers in your agent-aligned engineering standard.

Landmark requires Map's activity layer (Supabase). If your organization already set this up, Landmark works immediately. If not, ask your engineering leader to follow the Map setup guide. One command works without Supabase. marker reads directly from the YAML for your agent-aligned engineering standard.

Prerequisites

• Node.js 22+
• npm
• Map's activity layer up and populated (for most commands)
• A Landmark session created with fit-landmark login for every command except marker. See Sign In to Landmark for the full flow

Install

npm install @forwardimpact/landmark

Sign in

npx fit-landmark login --email you@example.com

login walks Supabase's magic-link flow and stores the session (0600) in your platform's config directory. On Linux that is ~/.config/landmark/credentials.json. See Sign In to Landmark for the macOS and Windows paths. Subsequent commands resolve your identity automatically.

Use --otp to skip the browser and paste the six-digit code instead. If your email is not in the activity roster, login fails. Your operator runs fit-terrain substrate provision to keep auth.users synchronized.

Browse marker definitions

Look up the observable indicators defined for any skill. The markers show what evidence each proficiency level expects:

npx fit-landmark marker data-integration

  Markers for Data Integration (data-integration)

    awareness:
      human:
        - Described differences between batch and streaming integrations.
        - Identified common source systems such as LIMS or ELN during onboarding.
      agent:
        - Summarized integration concepts and source systems accurately when prompted.

    foundational:
      human:
        - Implemented a simple field-to-field integration following an existing pattern.
        - Resolved a routine integration failure by inspecting logs and retrying.
      agent:
        - Produced a working integration scaffold using approved patterns.

Filter to a specific proficiency level:

npx fit-landmark marker data-integration --level working

Check your evidence

See which markers have evidence linked to your work:

npx fit-landmark evidence --email you@example.com

  Evidence

    code-review: 0 matched, 4 unmatched
      [unmatched] Reviewed moderate-complexity changes independently.
        rationale: Commit artifact; review behavior cannot be assessed.

    architecture-design: 1 matched, 2 unmatched
      [matched] Owned the architecture of a service through delivery and validation.
        rationale: A recent pull request introduced a new service boundary with documented API contracts.
      [unmatched] Evaluated alternatives and justified trade-offs in a written design review.
        rationale: No design review artifact present in this PR.

Each row shows the marker, whether it matched, the rationale for the assessment, and where the evidence came from. Each row also carries a provenance label. The label separates Map's direct artifact matches from Guide's deeper assessments. Guide evaluates your GitHub artifacts (PRs, reviews, commits) against each marker in the background. You see the results here. Filter by --skill to focus on a specific area.

View your skill coverage

See how complete your evidence record is across all expected skills:

npx fit-landmark coverage --email you@example.com

Coverage shows evidenced artifacts versus total expected markers. It gauges where your record is strong and where it has gaps.

Check promotion readiness

See which next-level markers you already evidenced and which are still outstanding:

npx fit-landmark readiness --email you@example.com

Readiness: you@example.com (J060 → J070)

    Architecture Design (practitioner):
      [ ] Led architecture for a multi-team initiative across regulated boundaries.
      [ ] Mentored engineers in producing high-quality, validated designs.
      [ ] Coordinated cross-system designs and surfaced integration risks proactively.

    Full Stack Development (practitioner):
      [ ] Led design and delivery of a complex application or major feature area.
      [ ] Defined component, API, and data access patterns adopted by the team.

Each unchecked marker has no evidence from you at the target level. Without --target, readiness checks against the next level above your current level. With --target, you can check against any specific level.

Track your growth timeline

See how your evidence accumulates over time, aggregated by quarter:

npx fit-landmark timeline --email you@example.com
npx fit-landmark timeline --email you@example.com --skill system_design

Timelines help you see whether growth accelerates, stalls, or concentrates in one area. Add --skill to focus on a specific capability.

Read your voice comments

See your own GetDX snapshot comments in a timeline view alongside evidence context:

npx fit-landmark voice --email you@example.com

All Landmark commands support --format text|json|markdown.

---

What's next