Conciseness

Nearly every line carries non-inferable operational data — measured token counts per mode, validated parser accuracy ("100% top-level symbol agreement" on 1,850 files), per-flag tradeoffs and cautions ("silently drops files whose only exports are consts"). It explains nothing Claude already knows and assumes competence, so every token earns its place rather than merely being efficient with minor trimmable spots.

Actionability

Fully executable guidance: the exact run command `node <this-skill-dir>/map.ts <repoRoot> [flags]`, every flag concretely defined with worked examples ("--include src/channels/turn maps exactly that subtree"), sizing tables for choosing flags, and the selftest command. Copy-paste ready commands cover the common cases including Swift-only and no-TypeScript environments.

Workflow Clarity

The Workflow guidance section sequences a clear two-tier process (dense whole-repo recon map, then exports/--members per-subsystem for the decision) with explicit validation checkpoints: "verify every dense-tier finding against typed signatures or source before acting", "spot-verify file paths and claims with grep before acting", and run selftest after editing map.ts. This matches the anchor for clear sequence with explicit validation steps; the destructive-operation cap does not apply to this read-only tool.

Progressive Disclosure

Well-organized single-file skill with clear section headers and zero nested or buried references (none of the referenced companion files exist as bundle files in this skill's directory). It falls short of the top anchor because the body runs past the simple-skill range and inlines deep-dive material — the Swift parser edge-case catalog and eleven sizing data points — that could be offloaded to a one-level-deep reference file to keep the always-loaded SKILL.md leaner. It is above the level-3 anchor since structure and placement are otherwise appropriate.

Specificity

The description lists multiple concrete actions and outputs — "Generate a single-file compressed symbol map", "files, exported symbols, typed signatures, plugin boundaries", "sized to fit an LLM context window" — with comprehensive coverage of what the skill produces. It matches the anchor for multiple specific concrete actions with no real gaps, so it is not the level below (minor gaps).

Completeness

It explicitly answers both "what" (generate a compressed symbol map with the listed contents) and "when" ("Use for whole-project reasoning: duplication hunting, refactor planning, architecture recon, or feeding another agent a full-project map") with concrete trigger phrases, in third-person voice. The missing-Use-when cap does not apply since explicit trigger guidance is present.

Trigger Term Quality

Natural phrases users would say are present — "duplication hunting", "refactor planning", "architecture recon", "whole-project reasoning", "TypeScript or Swift" — giving good keyword coverage. It falls short of the top anchor because common synonyms like "codebase overview", "repo map", or "code index" are missing.

Distinctiveness Conflict Risk

The skill occupies a clear niche — compressing a TS/Swift repo into a single context-window-sized symbol map — with triggers (whole-project reasoning, duplication hunting, feeding another agent) that are unlikely to fire for unrelated skills. Only trivial overlap exists with generic refactor-help phrasing, keeping conflict risk minimal.