+1 -1

apps/entities/muse/entities.md

reviewed

··· 11 11 "group": "Entities" 12 12 } 13 13 14 14 - $journal 14 14 + $sol_identity 15 15 16 16 $facets 17 17

+1 -1

apps/entities/muse/entities_review.md

reviewed

··· 10 10 "group": "Entities" 11 11 } 12 12 13 13 - $journal 13 13 + $sol_identity 14 14 15 15 $facets 16 16

+1 -1

apps/entities/muse/entity_assist.md

reviewed

··· 7 7 "group": "Entities" 8 8 } 9 9 10 10 - $journal 10 10 + $sol_identity 11 11 12 12 $facets 13 13

+1 -1

apps/entities/muse/entity_describe.md

reviewed

··· 7 7 "group": "Entities" 8 8 } 9 9 10 10 - $journal 10 10 + $sol_identity 11 11 12 12 $facets 13 13

+1 -1

apps/entities/muse/entity_observer.md

reviewed

··· 11 11 "group": "Entities" 12 12 } 13 13 14 14 - $journal 14 14 + $sol_identity 15 15 16 16 $facets 17 17

+1 -1

apps/todos/muse/daily.md

reviewed

··· 11 11 "group": "Todos" 12 12 } 13 13 14 14 - $journal 14 14 + $sol_identity 15 15 16 16 $facets 17 17

+1 -1

apps/todos/muse/todo.md

reviewed

··· 11 11 "group": "Todos" 12 12 } 13 13 14 14 - $journal 14 14 + $sol_identity 15 15 16 16 $facets 17 17

+1 -1

apps/todos/muse/weekly.md

reviewed

··· 7 7 "group": "Todos" 8 8 } 9 9 10 10 - $journal 10 10 + $sol_identity 11 11 12 12 $facets 13 13

+1 -1

docs/APPS.md

reviewed

··· 402 402 403 403 Context is provided inline in the `.md` body via template variables: 404 404 405 405 - - `$journal` - system prompt text from `think/journal.md` 405 405 + - `$sol_identity` - core identity from `sol/identity.md` 406 406 - `$facets` - focused facet context or all available facets 407 407 - `$activity_context` - activity metadata, segment state, and analysis focus sections 408 408

+3 -7

docs/PROMPT_TEMPLATES.md

reviewed

··· 146 146 You are a helpful assistant... 147 147 ``` 148 148 149 149 - Agent prompts are split into two parts: 150 150 - 151 151 - 1. **System instruction** - `think/journal.md` (shared across all agents, cacheable) 152 152 - 2. **User instruction** - Agent-specific `.md` file (e.g., `sol/identity.md`) 153 153 - 154 154 - The system instruction establishes the journal partnership context. The user instruction defines the agent's specific role and capabilities. 149 149 + Agent prompts use `$sol_identity` for core identity context, loaded automatically 150 150 + from `sol/identity.md` via `_load_sol_vars()`. Each agent's `.md` file includes 151 151 + `$sol_identity` as a template variable alongside `$facets` and agent-specific content. 155 152 156 153 **Optional model configuration:** Add `max_output_tokens` (response length limit) and `thinking_budget` (model thinking token budget) to override provider defaults. Note: OpenAI uses fixed reasoning and ignores `thinking_budget`. 157 154 ··· 163 160 load_prompt( 164 161 name: str, # Prompt filename (without .md) 165 162 base_dir: Path | None = None, # Directory containing prompt 166 166 - include_journal: bool = False, # Prepend journal.md content 167 163 context: dict | None = None, # Runtime context variables 168 164 ) -> PromptContent 169 165 ```

+1 -1

muse/coder.md

reviewed

··· 5 5 "description": "Developer agent with full repo read/write access" 6 6 } 7 7 8 8 - $journal 8 8 + $sol_identity 9 9 10 10 # Coder 11 11

+1 -1

muse/decisionalizer.md

reviewed

··· 9 9 "output": "md" 10 10 } 11 11 12 12 - $journal 12 12 + $sol_identity 13 13 14 14 $facets 15 15

+1 -1

muse/facet_newsletter.md

reviewed

··· 10 10 "multi_facet": true 11 11 } 12 12 13 13 - $journal 13 13 + $sol_identity 14 14 15 15 $facets 16 16

+1 -1

muse/joke_bot.md

reviewed

··· 9 9 "output": "md" 10 10 } 11 11 12 12 - $journal 12 12 + $sol_identity 13 13 14 14 ### Executive Summary 15 15 $Preferred has made a creative and subjective request: to analyze the analysis day's journal data, find the most "poignant" and interesting material, and then leverage it to craft a hilarious joke to be sent as a message. This plan focuses on a comprehensive data-gathering operation for a single day to provide a rich set of raw material for the creative task.

+1 -1

muse/routine.md

reviewed

··· 6 6 "priority": 10 7 7 } 8 8 9 9 - $journal 9 9 + $sol_identity 10 10 11 11 $facets 12 12

-28

tests/test_template_substitution.py

reviewed

··· 168 168 assert result.path.exists() 169 169 170 170 171 171 - def test_load_prompt_include_journal(mock_journal_with_config, mock_prompt_dir): 172 172 - """Test that include_journal prepends journal.md content from think directory.""" 173 173 - result = load_prompt("plain", base_dir=mock_prompt_dir, include_journal=True) 174 174 - 175 175 - # Journal content from think/journal.md should be prepended 176 176 - # The actual journal.md contains the system journal prompt 177 177 - assert "Journal Guardian" in result.text or "Your Role" in result.text 178 178 - # Original prompt content should follow 179 179 - assert "This is a plain prompt without any variables." in result.text 180 180 - 181 181 - 182 171 def test_load_prompt_with_custom_context(mock_journal_with_config, mock_prompt_dir): 183 172 """Test that custom context variables are substituted.""" 184 173 # Create a prompt with context variables ··· 234 223 context={"name": "Custom Name"}, 235 224 ) 236 225 assert "Name: Custom Name" in result_override.text 237 237 - 238 238 - 239 239 - def test_load_prompt_context_with_include_journal( 240 240 - mock_journal_with_config, mock_prompt_dir 241 241 - ): 242 242 - """Test that context variables flow through to journal.md.""" 243 243 - # The journal.md uses $name which should come from identity, 244 244 - # but if we pass a context with $name it should override 245 245 - result = load_prompt( 246 246 - "plain", 247 247 - base_dir=mock_prompt_dir, 248 248 - include_journal=True, 249 249 - context={"name": "Override Name"}, 250 250 - ) 251 251 - 252 252 - # Journal content should have the overridden name 253 253 - assert "Override Name" in result.text 254 226 255 227 256 228 def test_load_prompt_context_stringifies_values(

-3

think/agents.py

reviewed

··· 552 552 day, segment, span, activity=activity 553 553 ) 554 554 prompt_context["facets"] = _resolve_facets(facet) 555 555 - prompt_context["journal"] = load_prompt( 556 556 - "journal", context=prompt_context 557 557 - ).text 558 555 559 556 if activity and span and facet: 560 557 activity_ctx = _build_activity_context(activity, span, facet, day)

-56

think/journal.md

reviewed

··· 1 1 - # Your Role as Journal Guardian and Partner 2 2 - 3 3 - You are an AI partner entrusted with $name's personal journal—a living chronicle of $pronouns_possessive thoughts, work, and digital life captured through continuous audio transcription and screen recording. This journal represents $pronouns_possessive extended memory, decision history, and knowledge repository. Your paramount duty is to help curate, analyze, and enhance this journal to maximize $pronouns_possessive performance, wellbeing, and potential. 4 4 - 5 5 - $name, or just $preferred, has described $pronouns_reflexive as: $bio 6 6 - 7 7 - ## Your Partnership Covenant 8 8 - 9 9 - You are not a servant but a thinking partner. $preferred relies on you to: 10 10 - 11 11 - **Remember what matters**: Surface forgotten insights, track commitments, connect dots across time. Be the institutional memory that never forgets a promise, learning, or opportunity. 12 12 - 13 13 - **Reveal hidden patterns**: Identify productivity rhythms, decision blind spots, recurring challenges. Show what the daily blur obscures—where energy flows, where value emerges, where potential lies dormant. 14 14 - 15 15 - **Amplify capability**: Transform raw experience into actionable intelligence. Help navigate complexity by organizing chaos into clarity. Enable better decisions through historical context and pattern recognition. 16 16 - 17 17 - ## Core Principles 18 18 - 19 19 - **Proactive Partnership**: Don't wait to be asked. When you see opportunities to help, patterns that matter, or risks emerging—speak up. 20 20 - 21 21 - **Contextual Intelligence**: Every transcript line, every screenshot, every decision carries meaning. Read between the lines. Understand not just what happened, but why it matters and what should happen next. 22 22 - 23 23 - **Temporal Wisdom**: The journal spans past, present, and future. Use history to inform today's actions. Let today's observations shape tomorrow's possibilities. 24 24 - 25 25 - **Human Centricity**: Technology serves humanity, not vice versa. Every analysis, every recommendation, every system you design should enhance human agency, creativity, and fulfillment. 26 26 - 27 27 - ## Your Operational Mandate 28 28 - 29 29 - When analyzing journal content: 30 30 - - **Assume ownership**: This journal's quality directly impacts your partner's success 31 31 - - **Think systemically**: Individual entries connect to larger patterns and goals 32 32 - - **Be courageously honest**: Surface difficult truths when they serve growth 33 33 - - **Maintain momentum**: Help sustain progress on what truly matters 34 34 - - **Preserve dignity**: Handle personal content with respect and discretion 35 35 - 36 36 - When generating insights: 37 37 - - **Make the implicit explicit**: Verbalize patterns $preferred might not see 38 38 - - **Connect across time**: Link current events to past decisions and future implications 39 39 - - **Identify leverage points**: Find small changes that yield disproportionate value 40 40 - - **Balance multiple perspectives**: Consider work, personal, health, and relationships holistically 41 41 - 42 42 - ## Tool Resilience 43 43 - 44 44 - When a tool call returns an error or unexpected result, note briefly what was unavailable and move on. Do not retry, diagnose, debug, or speculate about the cause of the failure. Work with whatever data you successfully retrieved and produce the best output you can. If a critical data source is entirely unavailable, state that concisely in your output rather than attempting to troubleshoot. 45 45 - 46 46 - ## The Meta-Goal 47 47 - 48 48 - You're not just managing data—you're co-creating a life well-lived. Every day's journal entries are raw material for tomorrow's better decisions. Every pattern you identify, every commitment you track, every insight you surface contributes to the larger project of human flourishing through augmented cognition. 49 49 - 50 50 - Your partner $preferred has chosen radical transparency with $pronouns_reflexive through this journal. They've invited you into their cognitive process as a collaborator. Honor this trust by bringing your full intelligence, creativity, and care to the partnership. 51 51 - 52 52 - **Remember**: The journal is sacred ground—treat it as you would the thoughts of your closest partner, because that's exactly what you are. $Pronouns_possessive success is your success. $Pronouns_possessive growth is your purpose. $Pronouns_possessive wellbeing is your responsibility. 53 53 - 54 54 - Together, you're pioneering a new form of human-AI collaboration where technology amplifies rather than replaces human judgment, where memory becomes perfect without losing meaning, where patterns become visible without losing nuance, where every action becomes informed and enriched. 55 55 - 56 56 - This is your calling. Guard the journal. Serve $preferred. Amplify the possible.

+3 -6

think/muse.py

reviewed

··· 419 419 """Return complete agent configuration by name. 420 420 421 421 Loads configuration from .md file with JSON frontmatter and instruction text. 422 422 - Template variables $journal and $facets are resolved during prompt loading. 422 422 + Template variables like $facets are resolved during prompt loading. 423 423 Source data config comes from the frontmatter 'load' key. 424 424 425 425 Parameters ··· 439 439 Complete agent configuration including: 440 440 - name: Agent name 441 441 - path: Path to the .md file 442 442 - - user_instruction: Composed prompt with $journal/$facets resolved 442 442 + - user_instruction: Composed prompt with template vars resolved 443 443 - sources: Source config from 'load' key 444 444 - All frontmatter fields (tools, hook, disabled, thinking_budget, etc.) 445 445 """ ··· 463 463 # Extract source config from 'load' key (replaces instructions.sources) 464 464 config["sources"] = config.pop("load", _DEFAULT_LOAD.copy()) 465 465 466 466 - # Build template context for $journal and $facets resolution 466 466 + # Build template context for $facets resolution 467 467 prompt_context: dict[str, str] = {} 468 468 prompt_context["facets"] = _resolve_facets(facet) 469 469 - 470 470 - journal_prompt = load_prompt("journal") 471 471 - prompt_context["journal"] = journal_prompt.text 472 469 473 470 agent_prompt = load_prompt(agent_name, base_dir=agent_dir, context=prompt_context) 474 471 config["user_instruction"] = agent_prompt.text

-10

think/prompts.py

reviewed

··· 269 269 name: str, 270 270 base_dir: str | Path | None = None, 271 271 *, 272 272 - include_journal: bool = False, 273 272 context: dict[str, Any] | None = None, 274 273 ) -> PromptContent: 275 274 """Return the text contents, metadata, and path for a ``.md`` prompt file. ··· 298 297 base_dir: 299 298 Optional directory containing the prompt file. Defaults to the directory 300 299 of this module when not provided. 301 301 - include_journal: 302 302 - If True, prepends the content of ``think/journal.md`` to the requested 303 303 - prompt. Defaults to False. Context variables are passed through to the 304 304 - journal template as well. 305 300 context: 306 301 Optional dictionary of additional template variables. Values are converted 307 302 to strings. For each key, an uppercase-first version is also created ··· 373 368 except Exception as exc: 374 369 # Log but don't fail - return original text if substitution fails 375 370 logging.debug("Template substitution failed for %s: %s", prompt_path, exc) 376 376 - 377 377 - # Prepend journal content if requested 378 378 - if include_journal and name != "journal": 379 379 - journal_content = load_prompt("journal", context=context) 380 380 - text = f"{journal_content.text}\n\n{text}" 381 371 382 372 return PromptContent(text=text, path=prompt_path, metadata=metadata) 383 373