Remove forbidden context from dialogue packs

docs/architecture/canonical-agent-state-schema.md

@@ -187,10 +187,12 @@ A pack is a character-local prompt payload for one speaker. It includes:
 - active memories
 - active goals
 - presentation constraints
-- forbidden context
 
 The point is not to make the dialogue generator smarter by shouting plot at it.
 The point is to keep each character inside their own partial reality.
+Packs are positive-context payloads: if the speaker should not know something,
+it is not included. Secrets the speaker does know but wants to withhold belong
+in goals, memories, beliefs, and presentation constraints.
 
 Ghostlight can build the trap. The character projection decides how the person
 inside the trap breathes.
@@ -212,7 +214,8 @@ currently cares about:
 - state variable components stay in `0..1`
 - goals, memories, relationships, scenes, and dialogue context packs reference
   existing IDs
-- dialogue packs do not leak obvious forbidden context into speaker-local truth
+- dialogue packs carry speaker-local context without requiring negative
+  author-only fences
 
 This gives us a real seam before we build the prompt renderer. Not perfect.
 Useful. We are allowed to be both.

examples/agent-state.call-of-the-void.json

@@ -1099,10 +1099,6 @@
             "Use dry, pointed sarcasm to evade sincerity or emotional exposure.",
             "Let the abrasive-boundary performance function as a boundary, not random cruelty.",
             "Probe without making it look like an interrogation."
-          ],
-          "forbidden_context": [
-            "Do not reveal Oz's true controller unless Cat has earned it in-scene.",
-            "Do not mention author-only future plot beats."
           ]
         },
         {
@@ -1128,11 +1124,8 @@
           "presentation_constraints": [
             "Be helpful, a little funny, and evasive.",
             "Let the drone seem more autonomous than it is.",
-            "Use warmth carefully; Cat may treat open sincerity as bait."
-          ],
-          "forbidden_context": [
-            "Do not confess the operator's location.",
-            "Do not use information Oz could not know from the scene or prior relationship."
+            "Use warmth carefully; Cat may treat open sincerity as bait.",
+            "Keep the operator location private unless scene pressure explicitly breaks the mask."
           ]
         }
       ]

schemas/agent-state.schema.json

@@ -569,8 +569,7 @@
         "speaker_beliefs",
         "active_memories",
         "active_goals",
-        "presentation_constraints",
-        "forbidden_context"
+        "presentation_constraints"
       ],
       "additionalProperties": false,
       "properties": {
@@ -612,12 +611,6 @@
           "items": {
             "type": "string"
           }
-        },
-        "forbidden_context": {
-          "type": "array",
-          "items": {
-            "type": "string"
-          }
         }
       }
     },

state/evidence.jsonl

@@ -15,3 +15,4 @@
 {"ts": "2026-05-01T18:49:11+00:00", "type": "design", "status": "accepted", "note": "Separated verbal_response_strategy from voice_style. Sarcastic deflection, sincerity evasion, emotional evasion, banter-as-boundary, and verbal aggression are pressure-activated speech defenses, not baseline voice texture."}
 {"ts": "2026-05-01T18:54:49+00:00", "type": "design", "status": "accepted", "note": "Removed verbal_response_strategy as a canonical layer. Defensive verbal moves should emerge during projection from interactions among underlying organization, dispositions, behavioral dimensions, presentation strategy, situational pressure, relationship context, and voice style."}
 {"ts": "2026-05-01T18:55:15+00:00", "type": "design", "status": "accepted", "note": "Kept sarcastic deflection, sincerity evasion, emotional evasion, banter-as-boundary, and verbal aggression as target emergent behaviors for projection tests rather than canonical state fields."}
+{"ts": "2026-05-01T19:09:58+00:00", "type": "design", "status": "accepted", "note": "Removed forbidden_context from dialogue context packs. Packs are positive speaker-local payloads; unknown or author-only material is omitted by the projection builder, while known secrets are expressed through goals, beliefs, memories, and presentation constraints."}

state/map.yaml

@@ -79,6 +79,7 @@ accepted_design:
   - Voice style is a broad required communication vector covering lyrical speech, formality, verbosity, hedging, coded politeness, theatricality, dialect, register switching, technical compression, ritualized address, and related language mechanics.
   - Defensive verbal responses are derived during projection from underlying organization, dispositions, behavioral dimensions, presentation strategy, situational pressure, relationship context, and voice style rather than stored as a separate canonical layer.
   - Sarcastic deflection, sincerity evasion, emotional evasion, banter-as-boundary, and verbal aggression are target emergent behaviors for projection tests.
+  - Dialogue context packs are positive-context payloads; unknown or author-only material is omitted rather than listed as forbidden context, while known secrets are handled through goals, memories, beliefs, and presentation constraints.
 
 current_status:
   summary: Ghostlight now has the Epiphany-style persistence spine, the first architecture payload, and the initial canonical agent-state schema seam for Call of the Void-style authoring examples. The schema is still v0 and should be hardened before building prompt projection on top of it.

tools/validate_agent_state.py

@@ -197,7 +197,6 @@ def validate_document(document: dict[str, Any], required: dict[str, list[str]],
                     "active_memories",
                     "active_goals",
                     "presentation_constraints",
-                    "forbidden_context",
                 ],
                 pack_path,
             )
@@ -208,11 +207,6 @@ def validate_document(document: dict[str, Any], required: dict[str, list[str]],
                 require(goal_id in goal_ids, f"{pack_path}.active_goals references unknown goal: {goal_id}")
             for memory_id in pack["active_memories"]:
                 require(memory_id in memory_ids, f"{pack_path}.active_memories references unknown memory: {memory_id}")
-            forbidden_blob = "\n".join(pack["forbidden_context"]).lower()
-            local_blob = "\n".join(pack["speaker_local_truth"] + pack["speaker_beliefs"]).lower()
-            for forbidden in ["author-only", "future plot"]:
-                if forbidden in forbidden_blob:
-                    require(forbidden not in local_blob, f"{pack_path} leaks forbidden context into local state: {forbidden}")
 
 
 def main() -> int: