personal memory agent
CPO: cluster layer semantic rename hop 2 — screen → percepts
Rename the screen source type to percepts across the cluster layer,
muse configs, CLI, importers, stats, and all tests. The entry prefix
changes from "screen" to "percept" (singular, matching "transcript").
The rendering header "### Screen Activity" is unchanged — it describes
the content origin, not the source type. Agent filter dicts like
{"agents": {"screen": true}} are also unchanged as they reference
agent output filenames.
Also updates inline muse configs in test_generate_full.py and
test_output_hooks.py from old "audio"/"screen" keys to
"transcripts"/"percepts" (audio was stale since hop 1).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
+154
-137
+12
-10
apps/transcripts/call.py
+12
-10
apps/transcripts/call.py
···
46
46
else:
47
47
typer.echo(" (none)")
48
48
49
-
typer.echo("Screen:")
49
+
typer.echo("Percepts:")
50
50
if screen_ranges:
51
51
for start, end in screen_ranges:
52
52
typer.echo(f" {start} - {end}")
···
94
94
raw: bool = typer.Option(False, "--raw", help="Include transcripts and screen only."),
95
95
transcripts: bool = typer.Option(False, "--transcripts", help="Include transcript content."),
96
96
audio: bool = typer.Option(False, "--audio", help="Alias for --transcripts.", hidden=True),
97
-
screen: bool = typer.Option(False, "--screen", help="Include screen transcripts."),
97
+
percepts: bool = typer.Option(False, "--percepts", help="Include screen percepts."),
98
+
screen: bool = typer.Option(False, "--screen", help="Alias for --percepts.", hidden=True),
98
99
agents: bool = typer.Option(False, "--agents", help="Include agent outputs."),
99
100
max_bytes: int = typer.Option(
100
101
16384, "--max", help="Max output bytes (0 = unlimited)."
···
104
105
day = resolve_sol_day(day)
105
106
segment = resolve_sol_segment(segment)
106
107
stream = stream or get_sol_stream()
107
-
# --audio is an alias for --transcripts
108
+
# --audio is an alias for --transcripts, --screen is an alias for --percepts
108
109
transcripts = transcripts or audio
110
+
percepts = percepts or screen
109
111
110
112
if full and raw:
111
113
typer.echo("Error: Cannot use --full and --raw together.", err=True)
112
114
raise typer.Exit(1)
113
115
114
-
if (full or raw) and (transcripts or screen or agents):
116
+
if (full or raw) and (transcripts or percepts or agents):
115
117
typer.echo(
116
118
"Error: Cannot mix --full/--raw with individual source flags.", err=True
117
119
)
118
120
raise typer.Exit(1)
119
121
120
122
if full:
121
-
sources: dict[str, bool] = {"transcripts": True, "screen": True, "agents": True}
123
+
sources: dict[str, bool] = {"transcripts": True, "percepts": True, "agents": True}
122
124
elif raw:
123
-
sources = {"transcripts": True, "screen": True, "agents": False}
124
-
elif transcripts or screen or agents:
125
-
sources = {"transcripts": transcripts, "screen": screen, "agents": agents}
125
+
sources = {"transcripts": True, "percepts": True, "agents": False}
126
+
elif transcripts or percepts or agents:
127
+
sources = {"transcripts": transcripts, "percepts": percepts, "agents": agents}
126
128
else:
127
-
sources = {"transcripts": True, "screen": False, "agents": True}
129
+
sources = {"transcripts": True, "percepts": False, "agents": True}
128
130
129
131
if segment and (start or length is not None):
130
132
typer.echo("Error: Cannot mix --segment with --start/--length.", err=True)
···
158
160
transcript_ranges, screen_ranges = cluster_scan(day)
159
161
if transcript_ranges or screen_ranges:
160
162
days_with_data += 1
161
-
typer.echo(f"{day} transcripts:{len(transcript_ranges)} screen:{len(screen_ranges)}")
163
+
typer.echo(f"{day} transcripts:{len(transcript_ranges)} percepts:{len(screen_ranges)}")
162
164
163
165
if not days_with_data:
164
166
typer.echo(f"No data for {month}.")
+1
-1
apps/transcripts/tests/test_call.py
+1
-1
apps/transcripts/tests/test_call.py
···
15
15
result = runner.invoke(call_app, ["transcripts", "scan", "20240101"])
16
16
assert result.exit_code == 0
17
17
assert "Transcripts:" in result.output
18
-
assert "Screen:" in result.output
18
+
assert "Percepts:" in result.output
19
19
20
20
def test_scan_empty_day(self):
21
21
result = runner.invoke(call_app, ["transcripts", "scan", "20990101"])
+1
-1
muse/activities.md
+1
-1
muse/activities.md
+1
-1
muse/activity.md
+1
-1
muse/activity.md
+1
-1
muse/activity_state.md
+1
-1
muse/activity_state.md
+1
-1
muse/daily_schedule.md
+1
-1
muse/daily_schedule.md
+1
-1
muse/decisions.md
+1
-1
muse/decisions.md
+1
-1
muse/documentation.md
+1
-1
muse/documentation.md
+1
-1
muse/entities.md
+1
-1
muse/entities.md
+1
-1
muse/facets.md
+1
-1
muse/facets.md
+1
-1
muse/files.md
+1
-1
muse/files.md
+1
-1
muse/flow.md
+1
-1
muse/flow.md
+1
-1
muse/followups.md
+1
-1
muse/followups.md
+1
-1
muse/knowledge_graph.md
+1
-1
muse/knowledge_graph.md
+1
-1
muse/media.md
+1
-1
muse/media.md
+1
-1
muse/meetings.md
+1
-1
muse/meetings.md
+1
-1
muse/messaging.md
+1
-1
muse/messaging.md
+1
-1
muse/observation.md
+1
-1
muse/observation.md
+1
-1
muse/opportunities.md
+1
-1
muse/opportunities.md
+1
-1
muse/research.md
+1
-1
muse/research.md
+1
-1
muse/schedule.md
+1
-1
muse/schedule.md
+1
-1
muse/screen.md
+1
-1
muse/screen.md
+1
-1
muse/speakers.md
+1
-1
muse/speakers.md
+1
-1
muse/timeline.md
+1
-1
muse/timeline.md
+1
-1
muse/tools.md
+1
-1
muse/tools.md
+17
-17
tests/baselines/api/stats/stats.json
+17
-17
tests/baselines/api/stats/stats.json
···
12
12
"facets": false,
13
13
"sources": {
14
14
"agents": false,
15
-
"screen": false,
15
+
"percepts": false,
16
16
"transcripts": false
17
17
}
18
18
},
···
35
35
"facets": true,
36
36
"sources": {
37
37
"agents": false,
38
-
"screen": true,
38
+
"percepts": true,
39
39
"transcripts": true
40
40
}
41
41
},
···
59
59
"facets": true,
60
60
"sources": {
61
61
"agents": false,
62
-
"screen": true,
62
+
"percepts": true,
63
63
"transcripts": true
64
64
}
65
65
},
···
89
89
"agents": {
90
90
"screen": true
91
91
},
92
-
"screen": false,
92
+
"percepts": false,
93
93
"transcripts": true
94
94
}
95
95
},
···
120
120
"agents": {
121
121
"screen": true
122
122
},
123
-
"screen": false,
123
+
"percepts": false,
124
124
"transcripts": true
125
125
}
126
126
},
···
144
144
"facets": false,
145
145
"sources": {
146
146
"agents": false,
147
-
"screen": true,
147
+
"percepts": true,
148
148
"transcripts": true
149
149
}
150
150
},
···
166
166
"facets": true,
167
167
"sources": {
168
168
"agents": true,
169
-
"screen": false,
169
+
"percepts": false,
170
170
"transcripts": false
171
171
}
172
172
},
···
194
194
"agents": {
195
195
"screen": true
196
196
},
197
-
"screen": false,
197
+
"percepts": false,
198
198
"transcripts": true
199
199
}
200
200
},
···
224
224
"agents": {
225
225
"screen": true
226
226
},
227
-
"screen": false,
227
+
"percepts": false,
228
228
"transcripts": true
229
229
}
230
230
},
···
250
250
"agents": {
251
251
"screen": true
252
252
},
253
-
"screen": false,
253
+
"percepts": false,
254
254
"transcripts": true
255
255
}
256
256
},
···
280
280
"agents": {
281
281
"screen": true
282
282
},
283
-
"screen": false,
283
+
"percepts": false,
284
284
"transcripts": true
285
285
}
286
286
},
···
311
311
"agents": {
312
312
"screen": true
313
313
},
314
-
"screen": false,
314
+
"percepts": false,
315
315
"transcripts": true
316
316
}
317
317
},
···
335
335
"instructions": {
336
336
"sources": {
337
337
"agents": false,
338
-
"screen": true,
338
+
"percepts": true,
339
339
"transcripts": true
340
340
}
341
341
},
···
362
362
"agents": {
363
363
"screen": true
364
364
},
365
-
"screen": false,
365
+
"percepts": false,
366
366
"transcripts": true
367
367
}
368
368
},
···
381
381
"instructions": {
382
382
"sources": {
383
383
"agents": false,
384
-
"screen": "required",
384
+
"percepts": "required",
385
385
"transcripts": true
386
386
}
387
387
},
···
400
400
"instructions": {
401
401
"sources": {
402
402
"agents": false,
403
-
"screen": true,
403
+
"percepts": true,
404
404
"transcripts": "required"
405
405
}
406
406
},
···
424
424
"agents": {
425
425
"screen": true
426
426
},
427
-
"screen": false,
427
+
"percepts": false,
428
428
"transcripts": true
429
429
}
430
430
},
+16
-1
tests/baselines/api/todos/nudges.json
+16
-1
tests/baselines/api/todos/nudges.json
···
1
1
{
2
-
"nudges": []
2
+
"nudges": [
3
+
{
4
+
"day": "20260308",
5
+
"facet": "montague",
6
+
"index": 2,
7
+
"nudge": "20260309T09:00",
8
+
"text": "Recruit Benvolio for infrastructure support"
9
+
},
10
+
{
11
+
"day": "20260308",
12
+
"facet": "verona",
13
+
"index": 2,
14
+
"nudge": "20260309T09:00",
15
+
"text": "Prepare working demo for board meeting"
16
+
}
17
+
]
3
18
}
+3
-3
tests/baselines/api/transcripts/segments.json
+3
-3
tests/baselines/api/transcripts/segments.json
···
7
7
"stream": "default",
8
8
"types": [
9
9
"transcripts",
10
-
"screen"
10
+
"percepts"
11
11
]
12
12
},
13
13
{
···
17
17
"stream": "default",
18
18
"types": [
19
19
"transcripts",
20
-
"screen"
20
+
"percepts"
21
21
]
22
22
},
23
23
{
···
27
27
"stream": "default",
28
28
"types": [
29
29
"transcripts",
30
-
"screen"
30
+
"percepts"
31
31
]
32
32
}
33
33
]
+17
-17
tests/test_cluster.py
+17
-17
tests/test_cluster.py
···
25
25
"screen summary"
26
26
)
27
27
result, counts = mod.cluster(
28
-
"20240101", sources={"transcripts": True, "screen": False, "agents": True}
28
+
"20240101", sources={"transcripts": True, "percepts": False, "agents": True}
29
29
)
30
30
assert counts["transcripts"] == 1
31
31
assert counts["agents"] == 1
···
55
55
"20240101",
56
56
"120000",
57
57
"120100",
58
-
sources={"transcripts": True, "screen": False, "agents": True},
58
+
sources={"transcripts": True, "percepts": False, "agents": True},
59
59
)
60
60
# Check that the function works and includes expected sections
61
61
assert "### Transcript" in md
···
141
141
assert segments[1]["start"] == "10:00"
142
142
assert segments[1]["end"] == "10:10"
143
143
assert "transcripts" in segments[1]["types"]
144
-
assert "screen" in segments[1]["types"]
144
+
assert "percepts" in segments[1]["types"]
145
145
146
146
# Check third segment (screen only)
147
147
assert segments[2]["key"] == "110000_300"
148
148
assert segments[2]["start"] == "11:00"
149
149
assert segments[2]["end"] == "11:05"
150
-
assert segments[2]["types"] == ["screen"]
150
+
assert segments[2]["types"] == ["percepts"]
151
151
152
152
153
153
def test_cluster_period_uses_raw_screen(tmp_path, monkeypatch):
···
176
176
result, counts = mod.cluster_period(
177
177
"20240101",
178
178
"100000_300",
179
-
sources={"transcripts": True, "screen": True, "agents": False},
179
+
sources={"transcripts": True, "percepts": True, "agents": False},
180
180
)
181
181
182
182
# Should have both transcript and screen entries
183
183
assert counts["transcripts"] == 1
184
-
assert counts["screen"] == 1
184
+
assert counts["percepts"] == 1
185
185
assert "### Transcript" in result
186
186
# Should use raw screen format header
187
187
assert "Screen Activity" in result
···
218
218
"20240101",
219
219
"100000",
220
220
"100500",
221
-
sources={"transcripts": True, "screen": False, "agents": True},
221
+
sources={"transcripts": True, "percepts": False, "agents": True},
222
222
)
223
223
224
224
assert "### Transcript" in result
···
253
253
"20240101",
254
254
"100000",
255
255
"100500",
256
-
sources={"transcripts": False, "screen": True, "agents": False},
256
+
sources={"transcripts": False, "percepts": True, "agents": False},
257
257
)
258
258
259
259
assert "Screen Activity" in result
···
289
289
"20240101",
290
290
"100000",
291
291
"100500",
292
-
sources={"transcripts": False, "screen": True, "agents": False},
292
+
sources={"transcripts": False, "percepts": True, "agents": False},
293
293
)
294
294
295
295
# Should include content from both screen files
···
333
333
334
334
assert len(segments) == 1
335
335
assert segments[0]["key"] == "100000_300"
336
-
assert "screen" in segments[0]["types"]
336
+
assert "percepts" in segments[0]["types"]
337
337
338
338
339
339
def test_cluster_span(tmp_path, monkeypatch):
···
367
367
result, counts = mod.cluster_span(
368
368
"20240101",
369
369
["090000_300", "110000_300"],
370
-
sources={"transcripts": True, "screen": False, "agents": False},
370
+
sources={"transcripts": True, "percepts": False, "agents": False},
371
371
)
372
372
373
373
# Should have 2 transcript entries (one per segment)
374
374
assert counts["transcripts"] == 2
375
-
assert counts["screen"] == 0
375
+
assert counts["percepts"] == 0
376
376
assert "morning segment" in result
377
377
assert "late morning segment" in result
378
378
# Should NOT include the skipped segment
···
398
398
mod.cluster_span(
399
399
"20240101",
400
400
["090000_300", "100000_300"],
401
-
sources={"transcripts": True, "screen": False, "agents": False},
401
+
sources={"transcripts": True, "percepts": False, "agents": False},
402
402
)
403
403
404
404
assert "100000_300" in str(exc_info.value)
···
424
424
# Test filtering to only include entities
425
425
result, counts = mod.cluster(
426
426
"20240101",
427
-
sources={"transcripts": True, "screen": False, "agents": {"entities": True}},
427
+
sources={"transcripts": True, "percepts": False, "agents": {"entities": True}},
428
428
)
429
429
430
430
assert counts["transcripts"] == 1
···
455
455
"20240101",
456
456
sources={
457
457
"transcripts": True,
458
-
"screen": False,
458
+
"percepts": False,
459
459
"agents": {"entities": True, "meetings": "required", "flow": False},
460
460
},
461
461
)
···
488
488
"20240101",
489
489
sources={
490
490
"transcripts": True,
491
-
"screen": False,
491
+
"percepts": False,
492
492
"agents": {"entities": False, "todos:review": True},
493
493
},
494
494
)
···
515
515
# Empty dict should mean no agents
516
516
result, counts = mod.cluster(
517
517
"20240101",
518
-
sources={"transcripts": True, "screen": False, "agents": {}},
518
+
sources={"transcripts": True, "percepts": False, "agents": {}},
519
519
)
520
520
521
521
assert counts["transcripts"] == 1
+3
-3
tests/test_cluster_full.py
+3
-3
tests/test_cluster_full.py
···
29
29
copy_day(tmp_path)
30
30
monkeypatch.setenv("JOURNAL_PATH", str(tmp_path))
31
31
md, counts = mod.cluster(
32
-
"20240101", sources={"transcripts": True, "screen": False, "agents": True}
32
+
"20240101", sources={"transcripts": True, "percepts": False, "agents": True}
33
33
)
34
34
# Transcript entries come from 2 segments on 20240101 (default + import.apple)
35
35
assert counts["transcripts"] == 2
···
45
45
copy_day(tmp_path)
46
46
monkeypatch.setenv("JOURNAL_PATH", str(tmp_path))
47
47
out, _counts = mod.cluster(
48
-
"20240101", sources={"transcripts": True, "screen": False, "agents": True}
48
+
"20240101", sources={"transcripts": True, "percepts": False, "agents": True}
49
49
)
50
50
# Now uses insight format: "### {stem} summary"
51
51
assert "### screen summary" in out
···
59
59
"20240101",
60
60
"123456",
61
61
"123556",
62
-
sources={"transcripts": True, "screen": True, "agents": False},
62
+
sources={"transcripts": True, "percepts": True, "agents": False},
63
63
)
64
64
# Range mode with screen=True uses raw screen data.
65
65
assert "### Screen Activity" in out
+4
-4
tests/test_generate_full.py
+4
-4
tests/test_generate_full.py
···
79
79
80
80
test_generator = tmp_path / "test_gen.md"
81
81
test_generator.write_text(
82
-
'{\n "type": "generate",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nTest prompt'
82
+
'{\n "type": "generate",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nTest prompt'
83
83
)
84
84
85
85
# Mock the underlying generation function in think.models
···
146
146
147
147
test_generator = tmp_path / "hooked_gen.md"
148
148
test_generator.write_text(
149
-
'{\n "type": "generate",\n "title": "Hooked",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "test_hook"},\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nTest prompt'
149
+
'{\n "type": "generate",\n "title": "Hooked",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "test_hook"},\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nTest prompt'
150
150
)
151
151
152
152
# Mock the underlying generation function in think.models
···
198
198
199
199
test_generator = tmp_path / "nohook_gen.md"
200
200
test_generator.write_text(
201
-
'{\n "type": "generate",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nNo hook prompt'
201
+
'{\n "type": "generate",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nNo hook prompt'
202
202
)
203
203
204
204
# Mock the underlying generation function in think.models
···
265
265
266
266
test_generator = tmp_path / "empty_gen.md"
267
267
test_generator.write_text(
268
-
'{\n "type": "generate",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nTest prompt'
268
+
'{\n "type": "generate",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nTest prompt'
269
269
)
270
270
271
271
monkeypatch.setenv("GOOGLE_API_KEY", "x")
+1
-1
tests/test_generators.py
+1
-1
tests/test_generators.py
···
135
135
sources = instructions.get("sources", {})
136
136
137
137
assert sources.get("transcripts") == "required", "speakers should require transcripts"
138
-
assert sources.get("screen") is True, "speakers should include screen"
138
+
assert sources.get("percepts") is True, "speakers should include percepts"
139
139
140
140
141
141
def _write_temp_muse_prompt(stem: str, frontmatter: str) -> Path:
+1
-1
tests/test_journal_stats.py
+1
-1
tests/test_journal_stats.py
···
172
172
assert "token_usage_by_day" in data
173
173
assert "token_totals_by_model" in data
174
174
assert "total_transcript_duration" in data
175
-
assert "total_screen_duration" in data
175
+
assert "total_percept_duration" in data
176
176
assert (
177
177
data["token_usage_by_day"]["20240101"]["gemini-2.5-flash"]["total_tokens"]
178
178
== 495
+4
-4
tests/test_muse.py
+4
-4
tests/test_muse.py
···
39
39
40
40
def test_merge_instructions_config_sources_merge():
41
41
"""Test that sources dict is merged, not replaced."""
42
-
defaults = {"system": None, "sources": {"transcripts": False, "screen": False}}
42
+
defaults = {"system": None, "sources": {"transcripts": False, "percepts": False}}
43
43
overrides = {"sources": {"transcripts": True}}
44
44
result = _merge_instructions_config(defaults, overrides)
45
45
assert result["sources"]["transcripts"] is True # Overridden
46
-
assert result["sources"]["screen"] is False # Preserved from defaults
46
+
assert result["sources"]["percepts"] is False # Preserved from defaults
47
47
48
48
49
49
def test_merge_instructions_config_ignores_unknown_keys():
···
236
236
237
237
assert "sources" in result
238
238
assert result["sources"]["transcripts"] is False
239
-
assert result["sources"]["screen"] is False
239
+
assert result["sources"]["percepts"] is False
240
240
assert result["sources"]["agents"] is False
241
241
242
242
def test_sources_can_be_overridden(self, monkeypatch, tmp_path):
···
256
256
)
257
257
258
258
assert result["sources"]["transcripts"] is True # Overridden
259
-
assert result["sources"]["screen"] is False # Default preserved
259
+
assert result["sources"]["percepts"] is False # Default preserved
260
260
assert result["sources"]["agents"] is True # Overridden
261
261
262
262
+5
-5
tests/test_output_hooks.py
+5
-5
tests/test_output_hooks.py
···
170
170
171
171
prompt_file = tmp_path / "hooked_test.md"
172
172
prompt_file.write_text(
173
-
'{\n "type": "generate",\n "title": "Hooked",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "hooked_test"},\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nTest prompt'
173
+
'{\n "type": "generate",\n "title": "Hooked",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "hooked_test"},\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nTest prompt'
174
174
)
175
175
176
176
hook_file = tmp_path / "hooked_test.py"
···
224
224
225
225
prompt_file = tmp_path / "noop_test.md"
226
226
prompt_file.write_text(
227
-
'{\n "type": "generate",\n "title": "Noop",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "noop_test"},\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nTest prompt'
227
+
'{\n "type": "generate",\n "title": "Noop",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "noop_test"},\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nTest prompt'
228
228
)
229
229
230
230
hook_file = tmp_path / "noop_test.py"
···
270
270
271
271
prompt_file = tmp_path / "broken_test.md"
272
272
prompt_file.write_text(
273
-
'{\n "type": "generate",\n "title": "Broken",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "broken_test"},\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nTest prompt'
273
+
'{\n "type": "generate",\n "title": "Broken",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"post": "broken_test"},\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nTest prompt'
274
274
)
275
275
276
276
hook_file = tmp_path / "broken_test.py"
···
387
387
388
388
prompt_file = tmp_path / "prehooked_test.md"
389
389
prompt_file.write_text(
390
-
'{\n "type": "generate",\n "title": "Prehooked",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"pre": "prehooked_test"},\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nOriginal prompt'
390
+
'{\n "type": "generate",\n "title": "Prehooked",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"pre": "prehooked_test"},\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nOriginal prompt'
391
391
)
392
392
393
393
hook_file = tmp_path / "prehooked_test.py"
···
447
447
448
448
prompt_file = tmp_path / "both_hooks_test.md"
449
449
prompt_file.write_text(
450
-
'{\n "type": "generate",\n "title": "Both Hooks",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"pre": "both_hooks_test", "post": "both_hooks_test"},\n "instructions": {"system": "journal", "sources": {"audio": true, "screen": true}}\n}\n\nOriginal prompt'
450
+
'{\n "type": "generate",\n "title": "Both Hooks",\n "schedule": "daily",\n "priority": 10,\n "output": "md",\n "hook": {"pre": "both_hooks_test", "post": "both_hooks_test"},\n "instructions": {"system": "journal", "sources": {"transcripts": true, "percepts": true}}\n}\n\nOriginal prompt'
451
451
)
452
452
453
453
hook_file = tmp_path / "both_hooks_test.py"
+31
-31
think/cluster.py
+31
-31
think/cluster.py
···
94
94
segment_path: Path,
95
95
date_str: str,
96
96
transcripts: bool,
97
-
screen: bool,
97
+
percepts: bool,
98
98
agents: bool | dict[str, bool | str],
99
99
) -> list[dict[str, Any]]:
100
100
"""Process a single segment directory and return entries.
···
103
103
segment_path: Path to segment directory
104
104
date_str: Date in YYYYMMDD format
105
105
transcripts: Whether to load transcript content (JSONL and markdown)
106
-
screen: Whether to load raw screen data from *screen.jsonl files
106
+
percepts: Whether to load raw screen data from *screen.jsonl files
107
107
agents: Whether to load agent output summaries from *.md files.
108
108
Can be bool (all/none) or dict for selective filtering
109
109
(e.g., {"entities": True, "meetings": "required"}).
···
187
187
)
188
188
189
189
# Process raw screen data from screen.jsonl and *_screen.jsonl
190
-
if screen:
190
+
if percepts:
191
191
screen_files = list(segment_path.glob("*screen.jsonl"))
192
192
for screen_jsonl in screen_files:
193
193
try:
···
199
199
"segment_key": segment_key,
200
200
"segment_start": segment_start,
201
201
"segment_end": segment_end,
202
-
"prefix": "screen",
202
+
"prefix": "percept",
203
203
"content": content,
204
204
"name": f"{segment_path.name}/{screen_jsonl.name}",
205
205
"stream": stream,
···
254
254
255
255
256
256
def _load_entries(
257
-
day_dir: str, transcripts: bool, screen: bool, agents: bool | dict[str, bool | str]
257
+
day_dir: str, transcripts: bool, percepts: bool, agents: bool | dict[str, bool | str]
258
258
) -> list[dict[str, Any]]:
259
259
"""Load all transcript entries from a day directory."""
260
260
from think.utils import segment_parse
···
269
269
start_time, _ = segment_parse(seg_path.name)
270
270
if not start_time:
271
271
continue
272
-
entries.extend(_process_segment(seg_path, date_str, transcripts, screen, agents))
272
+
entries.extend(_process_segment(seg_path, date_str, transcripts, percepts, agents))
273
273
274
274
entries.sort(key=lambda e: e["timestamp"])
275
275
return entries
···
293
293
294
294
Maps the internal prefix names to source config names:
295
295
- "transcript" -> "transcripts"
296
-
- "screen" -> "screen"
296
+
- "percept" -> "percepts"
297
297
- "agent_output" -> "agents"
298
298
299
299
Returns:
300
-
Dict with counts for each source type, e.g., {"transcripts": 2, "screen": 1, "agents": 0}
300
+
Dict with counts for each source type, e.g., {"transcripts": 2, "percepts": 1, "agents": 0}
301
301
"""
302
302
# Map internal prefix to source config name
303
303
prefix_to_source = {
304
304
"transcript": "transcripts",
305
-
"screen": "screen",
305
+
"percept": "percepts",
306
306
"agent_output": "agents",
307
307
}
308
308
···
311
311
# Ensure all standard sources are present (even if 0)
312
312
return {
313
313
"transcripts": counts.get("transcripts", 0),
314
-
"screen": counts.get("screen", 0),
314
+
"percepts": counts.get("percepts", 0),
315
315
"agents": counts.get("agents", 0),
316
316
}
317
317
···
344
344
lines.append(f"### {header}")
345
345
lines.append(entry["content"].strip())
346
346
lines.append("")
347
-
elif entry["prefix"] == "screen":
347
+
elif entry["prefix"] == "percept":
348
348
lines.append("### Screen Activity")
349
349
lines.append(entry["content"].strip())
350
350
lines.append("")
···
407
407
408
408
date_str = _date_str(day_dir)
409
409
transcript_slots: set[datetime] = set()
410
-
screen_slots: set[datetime] = set()
410
+
percept_slots: set[datetime] = set()
411
411
day_path_obj = Path(day_dir)
412
412
413
413
# Check timestamp subdirectories for content files
···
437
437
if (seg_path / "screen.jsonl").exists() or any(
438
438
seg_path.glob("*_screen.jsonl")
439
439
):
440
-
screen_slots.add(slot)
440
+
percept_slots.add(slot)
441
441
442
442
transcript_ranges = _slots_to_ranges(sorted(transcript_slots))
443
-
screen_ranges = _slots_to_ranges(sorted(screen_slots))
444
-
return transcript_ranges, screen_ranges
443
+
percept_ranges = _slots_to_ranges(sorted(percept_slots))
444
+
return transcript_ranges, percept_ranges
445
445
446
446
447
447
def cluster_segments(day: str) -> list[dict[str, Any]]:
···
458
458
- key: segment directory name (HHMMSS_LEN format)
459
459
- start: start time as HH:MM
460
460
- end: end time as HH:MM
461
-
- types: list of content types present ("transcripts", "screen", or both)
461
+
- types: list of content types present ("transcripts", "percepts", or both)
462
462
"""
463
463
from think.utils import segment_parse
464
464
···
489
489
490
490
# Check for screen content
491
491
if (seg_path / "screen.jsonl").exists() or any(seg_path.glob("*_screen.jsonl")):
492
-
types.append("screen")
492
+
types.append("percepts")
493
493
494
494
if not types:
495
495
continue
···
546
546
547
547
Args:
548
548
day: Day in YYYYMMDD format
549
-
sources: Dict with keys "transcripts", "screen", "agents".
549
+
sources: Dict with keys "transcripts", "percepts", "agents".
550
550
Values can be bool, "required" string, or dict (for agents).
551
551
The "agents" source can be a dict for selective filtering,
552
552
e.g., {"entities": True, "meetings": "required"}.
553
553
554
554
Returns:
555
555
Tuple of (markdown, source_counts) where source_counts is a dict
556
-
with keys "transcripts", "screen", "agents" mapping to entry counts.
556
+
with keys "transcripts", "percepts", "agents" mapping to entry counts.
557
557
"""
558
558
empty_counts = {"transcripts": 0, "screen": 0, "agents": 0}
559
559
···
565
565
entries = _load_entries(
566
566
day_dir,
567
567
transcripts=sources.get("transcripts", False),
568
-
screen=sources.get("screen", False),
568
+
percepts=sources.get("percepts", False),
569
569
agents=sources.get("agents", False),
570
570
)
571
571
if not entries:
···
590
590
Args:
591
591
day: Day in YYYYMMDD format
592
592
segment: Segment key in HHMMSS_LEN format (e.g., "163045_300")
593
-
sources: Dict with keys "transcripts", "screen", "agents".
593
+
sources: Dict with keys "transcripts", "percepts", "agents".
594
594
Values can be bool, "required" string, or dict (for agents).
595
595
stream: Stream name. If None, searches all streams for the segment.
596
596
597
597
Returns:
598
598
Tuple of (markdown, source_counts) where source_counts is a dict
599
-
with keys "transcripts", "screen", "agents" mapping to entry counts.
599
+
with keys "transcripts", "percepts", "agents" mapping to entry counts.
600
600
"""
601
601
empty_counts = {"transcripts": 0, "screen": 0, "agents": 0}
602
602
···
608
608
entries = _load_entries_from_segment(
609
609
str(segment_dir),
610
610
transcripts=sources.get("transcripts", False),
611
-
screen=sources.get("screen", False),
611
+
percepts=sources.get("percepts", False),
612
612
agents=sources.get("agents", False),
613
613
)
614
614
if not entries:
···
622
622
def _load_entries_from_segment(
623
623
segment_dir: str,
624
624
transcripts: bool,
625
-
screen: bool,
625
+
percepts: bool,
626
626
agents: bool | dict[str, bool | str],
627
627
) -> list[dict[str, Any]]:
628
628
"""Load entries from a single segment directory.
···
630
630
Args:
631
631
segment_dir: Path to segment directory (e.g., /path/to/20251109/163045_300)
632
632
transcripts: Whether to load transcript content (JSONL and markdown)
633
-
screen: Whether to load raw screen data from *screen.jsonl files
633
+
percepts: Whether to load raw screen data from *screen.jsonl files
634
634
agents: Whether to load agent output summaries from *.md files
635
635
636
636
Returns:
···
639
639
segment_path_obj = Path(segment_dir)
640
640
# Parent is stream dir; grandparent is day dir
641
641
date_str = _date_str(str(segment_path_obj.parent.parent))
642
-
entries = _process_segment(segment_path_obj, date_str, transcripts, screen, agents)
642
+
entries = _process_segment(segment_path_obj, date_str, transcripts, percepts, agents)
643
643
entries.sort(key=lambda e: e["timestamp"])
644
644
return entries
645
645
···
660
660
Args:
661
661
day: Day in YYYYMMDD format
662
662
span: List of segment keys in HHMMSS_LEN format (e.g., ["163045_300", "170000_600"])
663
-
sources: Dict with keys "transcripts", "screen", "agents".
663
+
sources: Dict with keys "transcripts", "percepts", "agents".
664
664
Values can be bool, "required" string, or dict (for agents).
665
665
stream: Stream name. If None, searches all streams for each segment.
666
666
667
667
Returns:
668
668
Tuple of (markdown, source_counts) where source_counts is a dict
669
-
with keys "transcripts", "screen", "agents" mapping to entry counts.
669
+
with keys "transcripts", "percepts", "agents" mapping to entry counts.
670
670
671
671
Raises:
672
672
ValueError: If any segment directories are missing
···
692
692
segment_entries = _load_entries_from_segment(
693
693
str(seg_dir),
694
694
transcripts=sources.get("transcripts", False),
695
-
screen=sources.get("screen", False),
695
+
percepts=sources.get("percepts", False),
696
696
agents=sources.get("agents", False),
697
697
)
698
698
entries.extend(segment_entries)
···
735
735
day: Day in YYYYMMDD format
736
736
start: Start time in HHMMSS format
737
737
end: End time in HHMMSS format
738
-
sources: Dict with keys "transcripts", "screen", "agents".
738
+
sources: Dict with keys "transcripts", "percepts", "agents".
739
739
Values can be bool, "required" string, or dict (for agents).
740
740
"""
741
741
day_dir = str(day_path(day))
···
746
746
entries = _load_entries(
747
747
day_dir,
748
748
transcripts=sources.get("transcripts", False),
749
-
screen=sources.get("screen", False),
749
+
percepts=sources.get("percepts", False),
750
750
agents=sources.get("agents", False),
751
751
)
752
752
# Include segments that overlap with the requested range
+12
-12
think/journal_stats.py
+12
-12
think/journal_stats.py
···
23
23
self.days: Dict[str, Dict[str, float | int]] = {}
24
24
self.totals: Counter[str] = Counter()
25
25
self.total_transcript_duration = 0.0
26
-
self.total_screen_duration = 0.0
26
+
self.total_percept_duration = 0.0
27
27
self.agent_counts: Counter[str] = Counter()
28
28
self.agent_minutes: Counter[str] = Counter()
29
29
self.facet_counts: Counter[str] = Counter()
···
107
107
times_seconds = [self._parse_timestamp(t) for t in timestamps]
108
108
return max(times_seconds) - min(times_seconds)
109
109
110
-
def _calculate_screen_duration(self, frames: list) -> float:
110
+
def _calculate_percept_duration(self, frames: list) -> float:
111
111
"""Calculate screen duration from min/max frame timestamps."""
112
112
# Skip header (first element if it has no frame_id)
113
113
frame_timestamps = [
···
132
132
counts_for_totals = {
133
133
k: v
134
134
for k, v in stats.items()
135
-
if k not in ("transcript_duration", "screen_duration")
135
+
if k not in ("transcript_duration", "percept_duration")
136
136
}
137
137
self.totals.update(counts_for_totals)
138
138
139
139
# Accumulate durations
140
140
self.total_transcript_duration += stats.get("transcript_duration", 0.0)
141
-
self.total_screen_duration += stats.get("screen_duration", 0.0)
141
+
self.total_percept_duration += stats.get("percept_duration", 0.0)
142
142
143
143
# Apply agent data
144
144
day_agent_counts: Dict[str, int] = {}
···
175
175
"""Scan a single day and return stats dict for caching."""
176
176
stats: Counter[str] = Counter()
177
177
transcript_duration = 0.0
178
-
screen_duration = 0.0
178
+
percept_duration = 0.0
179
179
day_dir = Path(path)
180
180
181
181
# Track agent data for cache
···
225
225
screen_files = list(day_dir.glob("*/*/screen.jsonl"))
226
226
screen_files.extend(day_dir.glob("*/*/*_screen.jsonl"))
227
227
for jsonl_file in sorted(screen_files):
228
-
stats["screen_sessions"] += 1
228
+
stats["percept_sessions"] += 1
229
229
230
230
try:
231
231
frames = load_analysis_frames(jsonl_file)
···
235
235
236
236
# Count frames (excluding header)
237
237
frame_count = sum(1 for f in frames if "frame_id" in f)
238
-
stats["screen_frames"] += frame_count
238
+
stats["percept_frames"] += frame_count
239
239
240
240
# Calculate duration from timestamps
241
241
if frame_count > 0:
242
-
duration = self._calculate_screen_duration(frames)
243
-
screen_duration += duration
242
+
duration = self._calculate_percept_duration(frames)
243
+
percept_duration += duration
244
244
245
245
except (OSError, IOError) as e:
246
246
logger.warning(f"Error reading screen file {jsonl_file}: {e}")
···
328
328
329
329
# --- Build return dict ---
330
330
stats["transcript_duration"] = transcript_duration
331
-
stats["screen_duration"] = screen_duration
331
+
stats["percept_duration"] = percept_duration
332
332
333
333
return {
334
334
"stats": dict(stats),
···
460
460
logger.info(
461
461
f"Scanned {len(self.days)} days, "
462
462
f"{self.totals.get('transcript_sessions', 0)} transcript sessions, "
463
-
f"{self.totals.get('screen_sessions', 0)} screen sessions"
463
+
f"{self.totals.get('percept_sessions', 0)} percept sessions"
464
464
f"{cache_status}"
465
465
)
466
466
···
470
470
"days": self.days,
471
471
"totals": dict(self.totals),
472
472
"total_transcript_duration": self.total_transcript_duration,
473
-
"total_screen_duration": self.total_screen_duration,
473
+
"total_percept_duration": self.total_percept_duration,
474
474
"agent_counts": dict(self.agent_counts),
475
475
"agent_minutes": {k: round(v, 2) for k, v in self.agent_minutes.items()},
476
476
"agent_counts_by_day": self.agent_counts_by_day,
+3
-3
think/muse.py
+3
-3
think/muse.py
···
335
335
"activity": False,
336
336
"sources": {
337
337
"transcripts": False,
338
-
"screen": False,
338
+
"percepts": False,
339
339
"agents": False,
340
340
},
341
341
}
···
431
431
true = include current date/time in extra_context
432
432
- "day": false | true (default: false)
433
433
true = include analysis day context (requires analysis_day parameter)
434
-
- "sources": {"transcripts": bool, "screen": bool, "agents": bool|dict}
434
+
- "sources": {"transcripts": bool, "percepts": bool, "agents": bool|dict}
435
435
The "agents" source can be:
436
436
- bool: True (all agents), False (no agents)
437
437
- "required": all agents, fail if none found
···
445
445
- system_prompt_name: str - name of system prompt (for cache keys)
446
446
- user_instruction: str | None - loaded from user_prompt if provided
447
447
- extra_context: str | None - facets + now + day context
448
-
- sources: dict - {"transcripts": bool, "screen": bool, "agents": bool|dict}
448
+
- sources: dict - {"transcripts": bool, "percepts": bool, "agents": bool|dict}
449
449
"""
450
450
from think.utils import format_day
451
451