personal memory agent
Enable datetime context for agents and clean up backlog
Add instructions.now and instructions.day to agent configs that need
temporal awareness (entities, todos, newsletter, default chat, importer).
Update dream.py to always pass day in request config for both agents and
generators. Migrate preamble templates from $date to $day/$day_YYYYMMDD.
Remove completed backlog items.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
+14
-17
+1
-1
apps/entities/muse/entities_review.md
+1
-1
apps/entities/muse/entities_review.md
+1
-1
apps/entities/muse/entity_assist.md
+1
-1
apps/entities/muse/entity_assist.md
+1
-1
apps/entities/muse/entity_describe.md
+1
-1
apps/entities/muse/entity_describe.md
+1
-1
apps/todos/muse/weekly.md
+1
-1
apps/todos/muse/weekly.md
-3
docs/BACKLOG.md
-3
docs/BACKLOG.md
···
14
14
15
15
- [ ] Update supervisor/dream interaction to use dynamic daily schedule from daily schedule agent output
16
16
- [ ] Create segment agent for voiceprint detection and updating via hooks
17
-
- [ ] Refactor think/agents.py to use run_tools for all generation
18
-
- [ ] Evaluate moving cortex.py logic into agents.py for better separation of concerns
19
-
- [ ] Pass 'day' context through to tools in daily agents for correct storage location
20
17
- [ ] Surface named hook outputs in agents app and sol muse CLI
21
18
- [ ] Make daily schedule agents idempotent with state tracking (show existing vs new segments)
22
19
- [ ] Add activities attach/update MCP tools for facet curation (like entity tools)
+1
-1
muse/default.md
+1
-1
muse/default.md
+1
-1
muse/importer.md
+1
-1
muse/importer.md
+5
-5
think/dream.py
+5
-5
think/dream.py
···
255
255
256
256
logging.info(f"Spawning {prompt_name} for facet: {facet_name}")
257
257
258
-
request_config: dict = {"facet": facet_name}
258
+
# Always pass day for instructions.day context
259
+
request_config: dict = {"facet": facet_name, "day": day}
259
260
if is_generator:
260
-
request_config["day"] = day
261
261
request_config["output"] = config.get("output", "md")
262
262
if force:
263
263
request_config["force"] = True
···
284
284
# Regular single-instance prompt
285
285
logging.info(f"Spawning {prompt_name}")
286
286
287
-
request_config = {}
287
+
# Always pass day for instructions.day context
288
+
request_config: dict = {"day": day}
288
289
if is_generator:
289
-
request_config["day"] = day
290
290
request_config["output"] = config.get("output", "md")
291
291
if force:
292
292
request_config["force"] = True
···
303
303
agent_id = cortex_request(
304
304
prompt=prompt,
305
305
name=prompt_name,
306
-
config=request_config if request_config else None,
306
+
config=request_config,
307
307
)
308
308
spawned.append((agent_id, prompt_name, config))
309
309
logging.info(f"Started {prompt_name} (ID: {agent_id})")
+1
-1
think/templates/daily_preamble.md
+1
-1
think/templates/daily_preamble.md
···
1
-
You are an expert analyst tasked with analyzing $preferred's full workday transcript from **$date**. The transcript contains both audio conversations and screen activity data, organized into recording segments with timestamps.
1
+
You are an expert analyst tasked with analyzing $preferred's full workday transcript from **$day** ($day_YYYYMMDD). The transcript contains both audio conversations and screen activity data, organized into recording segments with timestamps.
2
2
3
3
You will be given the transcripts followed by a detailed request for how to process them. Follow those instructions carefully. Take time to consider all of the nuance of the interactions from the day, think through how best to prioritize the most important aspects, and formulate the best approach for each step of the analysis.
+1
-1
think/templates/segment_preamble.md
+1
-1
think/templates/segment_preamble.md
···
1
-
You are analyzing a recording segment from $preferred's workday on **$date**, covering **$segment_start to $segment_end**. This segment captures a specific time window of activity through audio transcription and screen recording.
1
+
You are analyzing a recording segment from $preferred's workday on **$day** ($day_YYYYMMDD), covering **$segment_start to $segment_end**. This segment captures a specific time window of activity through audio transcription and screen recording.
2
2
3
3
Focus your analysis on this discrete period - its context, activities, and significance within the broader day.