fit-summit - SKILL.md Agent Skill

name: fit-summit description: > Make staffing decisions you can defend by modeling team capability as a system. Use when a post-mortem surfaces the same skill gap again, when evaluating whether a hire, transfer, or promotion strengthens the team, when detecting structural risks like single points of failure, or when simulating what-if scenarios, aligning growth with team gaps, comparing teams, and tracking capability trajectory over time.

Summit

Team capability planning tool. Summit aggregates individual skill matrices into team-level views — coverage heatmaps, structural risks, and what-if scenarios. It treats a team as a system, not a collection of individuals: it measures what a team can do, not how well it is doing it.

When to Use

Understand what a team can do:

Viewing per-skill headcount depth — npx fit-summit coverage <team>
Detecting single points of failure, critical gaps, and concentration risks — npx fit-summit risks <team>
Overlaying evidence from Map's activity layer — npx fit-summit coverage <team> --evidenced

Make and defend staffing decisions:

Evaluating whether a hire strengthens the team — npx fit-summit what-if <team> --add '{ discipline: ..., level: ... }'
Simulating a departure or transfer — npx fit-summit what-if <team> --remove 'Name'
Comparing capability before and after a change — npx fit-summit what-if <team> --promote 'Name'
Evaluating project-specific coverage — npx fit-summit coverage <team> --project <name>

Growth and trajectory:

Growth opportunities aligned with team gaps — npx fit-summit growth <team>
Weighting recommendations by outcome scores — npx fit-summit growth <team> --outcomes
Tracking quarterly capability evolution — npx fit-summit trajectory <team>

Team comparison:

Diffing coverage and risks between teams — npx fit-summit compare <team1> <team2>
Reviewing roster composition — npx fit-summit roster

How It Works

Coverage

For each team member, Summit derives a skill matrix from their job (discipline, level, track) using the Map standard data. Skills at "working" proficiency or above count toward coverage depth. The result is a per-skill headcount showing how many people can meaningfully contribute to each skill area.

Structural Risks

Three risk types are detected from coverage data:

Single points of failure — skills with exactly one working+ holder. Severity depends on allocation: part-time holders are higher risk.
Critical gaps — skills expected by the team's disciplines and tracks that have zero working+ holders.
Concentration risks — clusters of 3+ people at the same level, capability, and proficiency, indicating lack of seniority distribution.

What-If Scenarios

Scenarios clone the roster, apply mutations (add/remove/move/promote), and diff coverage and risks against the original. The input is never modified — all simulation is pure.

Growth Alignment

Growth recommendations rank team members by their potential to fill team gaps. Each skill gap is classified by impact: critical > spof-reduction > coverage-strengthening. Candidates are ranked by current proficiency (lower is better — more room to grow) and level. When --outcomes is provided, recommendations within the same impact tier are re-sorted by worst GetDX driver score.

Evidence Decorator

The optional --evidenced flag loads evidence from Map's activity schema and overlays practiced capability onto derived coverage. This can escalate risks — a skill with derived depth but no evidence becomes a more urgent concern.

Trajectory

Trajectory reads the git history of the roster file, buckets commits by calendar quarter, and computes coverage at each point. Per-skill trends are classified as improving, declining, stable, or persistent_gap.

Audience Model

Each view applies privacy rules based on the audience:

Engineer — sees team aggregates and their own growth recommendations
Manager — sees individual-level detail within their team
Director — sees aggregated data with individual identity stripped

CLI Reference

See references/cli.md for full command listings.

Roster Format

See references/roster.md for YAML format and examples.

Common Workflows

See references/workflows.md for worked examples.

Prerequisites

Map standard data (from npx fit-map init)
A summit.yaml roster file (copy from the starter example)
Git repository (required for trajectory command)
Map activity layer (optional, for --evidenced and Map-sourced rosters)
GetDX integration (optional, for --outcomes)

Verification

npx fit-summit validate                   # Roster validates against agent-aligned engineering standard
npx fit-summit roster                     # Roster displays correctly
npx fit-summit coverage <team>            # Coverage heatmap renders
npx fit-summit risks <team>               # Risks detected as expected

Documentation

Summit Overview — Product overview, design principles, and audience model
Getting Started: Summit for Leaders — From zero to your first team capability analysis
Make Staffing Decisions You Can Defend — Coverage heatmaps, structural risks, and what-if scenarios
Evaluate a Candidate Against Team Gaps — Check whether a candidate fills the team's actual gap
Surface Capability Gaps — Detect single points of failure and coverage blindspots