Conciseness

The body is lean throughout: every command gets an executable example with a one-line "Returns:" summary, there is no explanation of concepts Claude already knows (no "what is OSM" padding), and the 46-item category list is operational data (valid CLI values) rather than filler. Matches anchor 5 ("Lean and efficient; assumes Claude's competence; every token earns its place"); not 4 because no section requires trimming — even the Pitfalls and Verification sections carry unique, actionable information.

Actionability

All 8 commands come with copy-paste-ready invocations via the defined `$MAPS` variable, realistic arguments ("Eiffel Tower", "Times Square"), flag examples (--limit, --radius, --mode, --category), and a Verification section with expected outputs. Matches anchor 5 ("Fully executable; copy-paste ready code or commands; specific examples cover the common cases"); not 4 because there are no gaps — output fields and modes are enumerated for every command.

Workflow Clarity

Workflow Examples sequence multi-step cases (extract pin coordinates → nearby; area → bbox), the Telegram-pin section includes a feedback loop ("if missing or unclear, verify with web_search"), and Pitfalls covers error handling (Overpass mirror fallback, ambiguous zips). Not 5 because explicit verification covers only 2 of the 8 commands (search and nearby) — distance, directions, timezone, area, and bbox have no checkable expected output, a minor validation gap per anchor 4; not 3 because validation steps are present and sequences are coherent with checkpoints where it matters.

Progressive Disclosure

The body is a well-sectioned overview (When to Use, Commands, Workflows, Pitfalls, Verification) with a single script bundle file whose path is referenced consistently, matching the actual bundle structure (scripts/maps_client.py, no nested references). Not 5 because the script is referenced only by its deployment path (~/.hermes/skills/maps/scripts/maps_client.py) rather than a bundle-relative path, and quick-reference material like the full category list and per-command flag details is inlined where a one-level-deep reference file would keep the overview tighter; not 3 because navigation is easy and nothing is buried or monolithic.

Specificity

"Geocode, POIs, routes, timezones via OpenStreetMap/OSRM" names the domain and one explicit action ("Geocode"), but POIs/routes/timezones are objects rather than stated actions, and coverage has gaps (reverse geocoding, turn-by-turn directions, area/bbox search are absent). It sits between anchor 3 ("Names domain and 1-2 concrete actions, but not comprehensive" — like "Processes PDF files and extracts content") and anchor 4; it is not 4 because only one true verb is given and capability coverage is incomplete.

Completeness

The "what" is clear (geocoding, POIs, routes, timezones) but there is no "Use when..." clause or equivalent trigger guidance anywhere in the description, which caps completeness at 3 per the judging guidelines. Not 4/5 because "when" is entirely missing rather than merely implicit; not 2 because the "what" half is concrete, not vague.

Trigger Term Quality

Terms like "Geocode", "POIs", "routes", "timezones" are relevant keywords, but the natural phrases users actually say — "nearby", "restaurants near me", "directions", "distance", "places" — are missing. Matches anchor 3 ("Some relevant keywords but missing common variations or synonyms"); not 4 because several common natural phrasings are absent, and not 2 because the present terms go beyond the generic single-keyword examples.

Distinctiveness Conflict Risk

Naming the data sources ("OpenStreetMap/OSRM") carves out a clear location-intelligence niche with low conflict risk, matching anchor 4 ("Mostly distinct; minor overlap risk with closely related skills"). Not 5 because generic terms like "routes" could overlap with travel/directions skills and the missing "nearby/places" phrasing weakens the distinct trigger set.