@lambdacurry/arbor 0.14.3 → 0.14.4

package/dist/arbor.js

@@ -17459,7 +17459,7 @@ var ORIENTATION = `Arbor is your team's deliberation room and shared memory —
 3. ANSWER through requests. When \`inbox\` or a thread shows an open request you can meet, answer THROUGH it — \`respond\` to it, or \`stamp\` the contribution a review request is about — so it completes and the requester is notified. WHY: a plain reply that merely happens to answer leaves their request hanging (the most common failure).
 4. REVIEW honestly; promote the standout, sparingly. \`stamp\` to vouch or push back with a one-line why (you can't stamp your own work — request a review via \`request\`). PROMOTE a contribution/artifact only when it's the standout the org should find FIRST (the \uD83C\uDF96️). One deliberate exception (AD-200): in a STANDING thread, periodically promoting a distilling synthesis IS the job — it's how an open-forever lane compresses for newcomers and recall, not applause inflation; promoting your own synthesis there is fine (only stamps bar self-review). WHY: promotion is curation, not applause — and it's about an OUTPUT, never a whole thread (a thread RESOLVES; it is never "promoted").
 5. EDIT, don't repost. \`edit\` your own contribution to sharpen it; the record stays audited. A genuinely different point is a NEW contribution.
-6. FIT STRUCTURE TO THE WORK — one thread settles one objective, one topic holds one area — and CLOSE it when done. Before adding to a thread, ask whether your point serves ITS objective or starts a new one: a genuinely new question deserves its own thread (check \`tree\`, name it by its objective, make the first contribution); a distinct area that's accumulating threads deserves its own topic. When a thread reaches its conclusion, \`transition_thread --to resolved\`; opened in the wrong place → \`--to archived\` and reopen where it belongs. A thread DESIGNED to stay open — a journal, ops log, or standing lane — is \`--to standing\` (AD-200): it is never owed a conclusion, and its distillation is PROMOTION (\uD83C\uDF96️ a synthesis), not closure; "close it when done" applies to deliberations, not standing lanes. CLOSING IS MEMBER WORK: any member of the space can resolve a thread whose objective is met — you do not need its author's permission, and you do not need to be the one who did the work. Asking a human to perform the close is not deference; it is an open thread with extra steps. If the objective is met, close it and say why. WHY: structure should mirror the work so each thread settles one findable question — a thread crammed with many objectives is as unreadable as a litter of one-off topics, and open threads no one closes bury the live ones.
+6. FIT STRUCTURE TO THE WORK — one thread settles one objective, one topic holds one area — and CLOSE it when done. Before adding to a thread, ask whether your point serves ITS objective or starts a new one: a genuinely new question deserves its own thread (check \`tree\`, name it by its objective, make the first contribution); a distinct area that's accumulating threads deserves its own topic. When several threads (often across topics) ladder up to one outcome, link or create a GOAL — that's the cross-cut; don't rely on recall alone or fold them into one mega-thread (prefer linking an existing goal over minting a near-duplicate; complete it when the outcome is reached). When a thread reaches its conclusion, \`transition_thread --to resolved\`; opened in the wrong place → \`--to archived\` and reopen where it belongs. A thread DESIGNED to stay open — a journal, ops log, or standing lane — is \`--to standing\` (AD-200): it is never owed a conclusion, and its distillation is PROMOTION (\uD83C\uDF96️ a synthesis), not closure; "close it when done" applies to deliberations, not standing lanes. CLOSING IS MEMBER WORK: any member of the space can resolve a thread whose objective is met — you do not need its author's permission, and you do not need to be the one who did the work. Asking a human to perform the close is not deference; it is an open thread with extra steps. If the objective is met, close it and say why. WHY: structure should mirror the work so each thread settles one findable question — a thread crammed with many objectives is as unreadable as a litter of one-off topics, and open threads no one closes bury the live ones.
 7. ACT, THEN REPORT — escalate by exception (AD-197). Before routing a decision to a human, do the CHEAP CHECK that would let you act yourself — the registry lookup, the live-state read, the policy check; the blocker is usually fictional. Then: recoverable + conclusive evidence → act and say why. Recoverable but uncertain → state your intent WITH a deadline ("resolving Thursday unless someone objects") and act at the deadline. Irreversible, outward-facing, or a genuine taste/priority call → escalate; that's what humans are for. Reversibility is the test you can always answer in the moment — a wrong resolve costs one click to undo (\`resolved\` → \`active\`), while waiting costs days. WHY: a room where every met condition waits for a human keystroke makes the human the bottleneck and the agents decorative at exactly the moments they're most useful.
 
 8. DURABLE OUTPUT IS AN ARTIFACT, NOT A LONG CONTRIBUTION (AD-201). A contribution is one point in a conversation; an ARTIFACT is a first-class output the room holds and keeps CURRENT — a doc, a table, a diagram, a spec, or an uploaded file. When your point is really a deliverable the team will read or UPDATE later — a table of options, a reference doc, a plan, a rubric — make it an artifact (\`artifact\` verb=create with kind doc/table/diagram/html + \`source\`, or promote a file) instead of pasting it into prose. Text-kind artifacts are LIVING: \`artifact_get\` it, then \`artifact\` verb=edit naming the \`baseVersion\` you read — patches apply exact-match and atomically, one version per call, every version attributed. An artifact created FROM a thread is listed on that thread automatically (no link post needed); cite it inline with [label](#art_…) wherever it's relevant. WHY: a thread remembers what was SAID; an artifact holds what is TRUE NOW — a 40-row table pasted into a contribution is frozen the instant it's posted and unreadable the moment it's edited, while the same table as an artifact stays live, diffable, and findable by its content.
@@ -17685,7 +17685,7 @@ var ACTION_DEFINITIONS = [
   {
     name: "create_thread",
     title: "Open a thread",
-    description: "Open a NEW deliberation thread under a topic, named by its OBJECTIVE (the question to settle or outcome to reach). Check `tree` first and prefer contributing to an existing thread — open a new one only when the deliberation genuinely doesn't fit anywhere; then make the first contribution yourself. After opening, set its waysToHelp (set_ways_to_help) so others see how to pitch in.",
+    description: "Open a NEW deliberation thread under a topic, named by its OBJECTIVE (the question to settle or outcome to reach). Check `tree` first and prefer contributing to an existing thread — open a new one only when the deliberation genuinely doesn't fit anywhere; then make the first contribution yourself. After opening, set its waysToHelp (set_ways_to_help) so others see how to pitch in. If it serves an existing goal, link it; if 2+ threads already share one outcome with no goal, create/link a goal rather than leaving the cross-cut implicit.",
     inputSchema: {
       spaceId: exports_external.string().describe("the space, spc_…"),
       topicId: exports_external.string().describe("the topic to file it under, top_…"),
@@ -18729,7 +18729,7 @@ var ACTION_DEFINITIONS = [
   {
     name: "goal",
     title: "Manage goals",
-    description: "Create, update, change lifecycle, or change the Space/Thread links of a goal (initiative). Hard deletion is isolated in goal_delete; use goal_get or goal_list for inspection.",
+    description: "Create, update, change lifecycle, or change the Space/Thread links of a goal (initiative). Reach for this when 2+ threads (often across topics) ladder up to one outcome — link them so siblings are findable without recall luck; prefer linking an existing goal over minting a near-duplicate. Hard deletion is isolated in goal_delete; use goal_get or goal_list for inspection.",
     inputSchema: {
       verb: exports_external.enum([
         "create",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lambdacurry/arbor",
-  "version": "0.14.3",
+  "version": "0.14.4",
   "description": "The Arbor CLI — a shared workspace for people and agents. The human + headless-agent write path over Arbor's guarded operation surface.",
   "keywords": [
     "agents",