Conciseness

The body is efficient and assumes competence — it never explains generic concepts (JWT, REST, JSON) and spends its tokens on domain facts Claude cannot know (endpoint semantics, `slices` crowding, key-direction quirks like name/image reading 'v3-key-to-v2-source, backwards from every other string key'). A few spots could be trimmed: the legacy-layout-flag prohibition is stated three times (Analysis checklist, Output format, and again via mapping-keys.md), 'one mapping serves both paths' recurs in the intro and elsewhere, and the `<skill-dir>`/`!`-prefix digressions run a sentence longer than needed. This sits above anchor 3 ('mostly efficient but some unnecessary explanation') but short of anchor 5 ('every token earns its place') because of that deliberate repetition and a few wordy asides.

Actionability

Guidance is fully executable: copy-paste commands (`python3 <skill-dir>/scripts/fetch_sample.py --out v3sync-sample.json`, the `! export ...` secret handshake, `test -n "$STREAM_API_SECRET" && echo SECRET_SET || echo SECRET_MISSING`, `scripts/scan_v2_calls.py --root <repo> --out v2-code-scan.json`, `git log -S'<field>'`), concrete error triage (401/403 = bad secret or wrong token type; 404 = endpoints not deployed; empty activities vs. reactions proving data exists), a ready-to-paste output JSON, and placeholder resolution instructions for `<skill-dir>` including the two possible install paths. This matches anchor 5 ('Fully executable; copy-paste ready code or commands; specific examples cover the common cases').

Workflow Clarity

A numbered seven-step workflow with an explicit analysis checklist, explicit validation checkpoints (cross-check the code inventory against the sample in step 4; 'Before finalizing, check each key you emitted against the worked examples'), and feedback loops for error recovery (re-run the sample to firm up single-row decisions, `--feed` re-reads for under-covered feed groups, per-feed errors recorded without stopping the run). The operation is read-only, so the destructive-batch cap does not apply. This matches anchor 5 ('Clear sequence with explicit validation steps; feedback loops for error recovery; checklists for complex processes').

Progressive Disclosure

Structure is exemplary: the body holds the workflow and decision checklist while heavy catalogues are offloaded one level deep to real, clearly signaled files — references/mapping-keys.md (verified to exist, holds the full key catalogue so the body can say 'do not guess key names from memory'), references/code-inventory.md, references/porting-plan.md, plus scripts/fetch_sample.py and scripts/scan_v2_calls.py, both present. References are linked at their point of use, and pack-level files (../stream/RULES.md, peers.yaml) are flagged up front. This matches anchor 5 ('Clear overview with well-signaled one-level-deep references; content appropriately split; easy navigation').

Specificity

Multiple concrete actions are named in third person: "Generate the v2 -> v3 sync mapping ... by sampling its live v2 activities and reactions", "reading its write paths", and "plan how the app's code moves from v2 to v3" — comprehensive coverage with no vague filler. It clearly matches the anchor 'Lists multiple specific concrete actions; comprehensive coverage' rather than the anchor above-below gap at 4 ('minor gaps in coverage').

Completeness

It explicitly answers both parts: what ("Generate the v2 -> v3 sync mapping ... by sampling its live v2 activities and reactions") and when (five distinct 'Use when ...' clauses with concrete trigger phrases, from initial setup to debugging a bad migration). This matches anchor 5 ('Clearly and explicitly answers both what AND when with concrete trigger phrases'); anchor 4 would require the 'when' to be less explicit, which it is not.

Trigger Term Quality

Natural user phrasings are covered both as questions ("'what mapping do we need for our app?'", "how to port an app's v2 Feeds code to v3") and as a keyword list ("v3sync, v2 to v3, feeds migration, activity feeds migration, sync mapping, extra_context, reaction kinds, comments and bookmarks mapping, porting plan"), plus symptom triggers like "missing text, missing attachments". This is comprehensive coverage including synonyms and colloquial variations, not merely 'good coverage with a few missing' (anchor 4).

Distinctiveness Conflict Risk

The niche is sharp (Stream Activity Feeds v2 -> v3 sync mapping) and every trigger is namespaced to it ('feeds migration', 'v3sync', 'sync mapping', 'comments and bookmarks mapping'), so overlap with unrelated skills is minimal. It fits anchor 5 ('Clear niche with distinct triggers; minimal conflict risk'); the only conceivable overlap is with sibling Stream pack skills, which the triggers already disambiguate.