@drawbridge/drawbridge-agents 0.1.49 → 0.1.50

package/claude/CLAUDE.md

@@ -14,6 +14,7 @@
 @../conventions/app-web-forms.md
 @../conventions/app-web-gates.md
 @../conventions/docs-linkage.md
+@../conventions/humanized-copy.md
 @../conventions/superpowers-docs.md
 @../conventions/git-branching.md
 @../conventions/graphify.md

package/conventions/humanized-copy.md

@@ -0,0 +1,25 @@
+# Humanize every word a person will read
+
+Any prose you write that a human will read as *product or communication* goes through the
+`humanizer` skill before it ships or posts. No exceptions for length — a two-line Asana
+comment counts.
+
+Covers, at minimum:
+
+- App/UI copy: labels, buttons, empty states, error and notice messages, tooltips, onboarding.
+- Asana comments and task descriptions, ship handoffs, retest instructions.
+- Docs and user stories, README prose, release notes.
+- Emails, SMS bodies, notification text, outreach drafts (growth), article/content drafts.
+- Anything else a customer, teammate, or prospect reads as writing.
+
+Does NOT cover: code, commit messages, code comments, test names, log lines, schema
+descriptions — engineer-facing text follows the code conventions instead.
+
+Rules when running it:
+
+- Run humanizer BEFORE posting/committing the copy, not after someone flags it.
+- It must not change facts, numbers, steps, or expected results — if a pass makes concrete
+  instructions vaguer, keep the original wording for those parts.
+- Leave markup intact (Asana mention tags, links, placeholders like `{name}`).
+- If the humanizer skill is not available in your session, say so and flag the copy as
+  un-humanized — do not silently skip the pass.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@drawbridge/drawbridge-agents",
-  "version": "0.1.49",
+  "version": "0.1.50",
   "description": "Shared agent-instruction content (rules, code style, conventions) for the drawbridge-* monorepo.",
   "license": "UNLICENSED",
   "publishConfig": {