Self-evolving second brain with 35 AI skills, 10 agents, and people CRM. Closed-loop harness: a V-model verification lifecycle where the worker never grades its own homework. Plus paired anti-slop design skills for marketing and product UI. Works with Claude Code, Cursor, Kiro, Gemini CLI, Codex.
Python agent-loopagent-pluginsagent-skillsagenticai-agentsclaude-codeclosed-loopcursorgarry-tangbraingstackknowledge-managementloop-engineeringobsidianpeople-crmproductivitysecond-brainself-evolvingverification-harnessworker-agents
fix: move the daily-journal trigger onto an always-loaded surface (#34)
Huy Tieudc5aeb647490
diff --git a/.claude-plugin/plugin.json b/.claude-plugin/plugin.jsonindex 363ba1a..18e81ec 100644--- a/.claude-plugin/plugin.json+++ b/.claude-plugin/plugin.json@@ -1,6 +1,6 @@{"name": "cog-second-brain",- "version": "3.10.0",+ "version": "3.10.1","description": "Self-evolving second brain with 33 AI agent skills, 10 agents, and people CRM. Closed-loop harness release: a V-model verification lifecycle (spec to evidence, criterion-traced checkpoints), read-only verifier agents that observe artifacts instead of trusting worker summaries, risk lanes, multi-session ultragoals with a north-star acceptance gate, plus craft skills for writing, visual work, and a paired anti-slop design system for marketing and product UI. Inspired by Garry Tan's gstack specialist sessions, gbrain knowledge patterns, and dwarves-kit closed-loop execution. Works with Claude Code, Antigravity, Cursor, Kiro, Gemini CLI, Codex, and any markdown-reading agent.","author": {"name": "Huy Tieu",diff --git a/.claude/skills/daily-journal/SKILL.md b/.claude/skills/daily-journal/SKILL.mdindex 6f895ca..a8913b4 100644--- a/.claude/skills/daily-journal/SKILL.md+++ b/.claude/skills/daily-journal/SKILL.md@@ -6,7 +6,7 @@ description: A passive daily work journal that Claude keeps FOR you so you never# Daily Journal## Purpose-Keep an **automatic, passive** journal of your workdays so you don't have to write one. Claude logs what got done as it happens; you optionally run a reflection later. This is distinct from `/daily-checkin` (which is a manual reflection where you supply the input) — here **Claude is the author** and the log accrues in the background.+Keep an **automatic, passive** journal of your workdays so you don't have to write one. Claude logs what got done as it happens; you optionally run a reflection later. This is distinct from `/weekly-checkin` (which is a manual reflection where you supply the input) — here **Claude is the author** and the log accrues in the background.Two modes:1. **`log`** (default, mostly invoked implicitly) — append a short entry to today's journal.@@ -17,7 +17,9 @@ Two modes:- Create the file from the template below on the first entry of the day.- Append only — never rewrite earlier entries. Newest entries go at the bottom of the Log section.-## When to log (behavioral trigger — applies in ANY session, not just when invoked)+## When to log (procedure for the always-apply trigger)+The trigger itself lives in `CLAUDE.md` § Daily Journal (ALWAYS APPLY), because a skill body is lazily loaded and cannot fire itself. This section is the procedure that trigger runs.+After finishing a **meaningful** unit of work, append one entry. Meaningful = would matter to future-you or shows how the day was spent. Examples:- Shipped/committed something, published a note/brief/slide, filed or moved tracker issues (Linear/Jira/GitHub).- Made a decision, changed direction, or hit a notable blocker.diff --git a/.cursor-plugin/plugin.json b/.cursor-plugin/plugin.jsonindex 6499309..cce0853 100644--- a/.cursor-plugin/plugin.json+++ b/.cursor-plugin/plugin.json@@ -1,7 +1,7 @@{"name": "cog-second-brain","displayName": "COG: Agentic Second Brain",- "version": "3.10.0",+ "version": "3.10.1","description": "Self-evolving second brain powered by AI agents, Obsidian markdown, and Git. 33 skills, 10 agents (6 workers + 4 read-only verifiers), people CRM with tiered enrichment. Closed-loop harness release: V-model verification lifecycle with criterion-traced evidence, risk lanes, multi-session ultragoals with a north-star acceptance gate, craft skills for writing and visuals, and paired anti-slop design skills for marketing and product UI. Works with Claude Code, Cursor, Kiro, Gemini CLI, and Codex.","author": {"name": "Huy Tieu",diff --git a/.cursorrules b/.cursorrulesindex d18eba4..698868a 100644--- a/.cursorrules+++ b/.cursorrules@@ -25,6 +25,10 @@ Before answering questions about people, projects, strategy, or decisions:2. If project-specific, read `04-projects/<project>/`3. Then synthesize an answer+## Daily Journal++Ambient, not a command. After finishing a meaningful unit of work (shipped or committed something, produced a deliverable, made a decision, hit a blocker), append one entry to `01-daily/journal/YYYY-MM-DD.md` (`date +%F`). Create the file from the template in `.claude/skills/daily-journal/SKILL.md` on the first entry of the day; append only. Skip trivia and scratch work, and do not announce the write.+## Integration PreferencesCheck `00-inbox/MY-INTEGRATIONS.md` before using any external integration:diff --git a/01-daily/journal/.gitkeep b/01-daily/journal/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/AGENTS.md b/AGENTS.mdindex d8ddbba..0fed850 100644--- a/AGENTS.md+++ b/AGENTS.md@@ -701,7 +701,7 @@ The following 7 skills raise output quality on writing and visual work. They enc**Triggers:**- `/daily-journal` or `/daily-journal reflect [today|yesterday|YYYY-MM-DD]`- "log this to my journal"-- Implicitly, after finishing a meaningful chunk of work in any session+- Implicitly, after finishing a meaningful chunk of work in any session. The always-apply trigger lives in `CLAUDE.md` § Daily Journal so it is loaded in every session; this entry is the reference.**Purpose:** Distinct from `/weekly-checkin`, where you supply the input. Here **the agent is the author** and the log accrues in the background, so the record exists even on days you'd never sit down to write one.diff --git a/CHANGELOG.md b/CHANGELOG.mdindex 92b75be..b706fbd 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -2,6 +2,25 @@All notable changes to COG (Cognition + Obsidian + Git) will be documented in this file.+## [3.10.1] - 2026-08-18++### Fixed++#### The daily journal is actually ambient now+`daily-journal` described itself as an automatic, passive log the agent keeps for you, but it logged nothing until you invoked it. The cause was architectural, not a missing instruction: the behavioral trigger ("append after finishing a meaningful unit of work") lived inside `SKILL.md`, and a skill body is lazily loaded, so the instruction telling the agent to act ambiently was itself locked behind the manual trigger.++A behavioral trigger cannot live in a lazily loaded file. The trigger now lives on the always-loaded surfaces and the skill body keeps the procedure.++- **`CLAUDE.md`** gains a `## Daily Journal (ALWAYS APPLY)` section carrying the trigger, the do-not-log list, and an explicit opt-out.+- **`.cursorrules`** gains the same rule in its own register, since Cursor does not read `CLAUDE.md`.+- **`AGENTS.md`** names `CLAUDE.md` as the trigger's home so the surfaces agree.+- **`SKILL.md`** stops claiming to be its own trigger and points at where the trigger actually lives.+- `01-daily/journal/` now ships with a `.gitkeep` so the destination exists on a fresh clone.++Also fixes a dangling reference in `daily-journal`'s purpose line: it compared itself to `/daily-checkin`, which this repo does not ship. The skill is `/weekly-checkin`.++Reported in #26, with the correct diagnosis.+## [3.10.0] - 2026-08-07### Addeddiff --git a/CLAUDE.md b/CLAUDE.mdindex 9baf37e..d2f5204 100644--- a/CLAUDE.md+++ b/CLAUDE.md@@ -208,6 +208,18 @@ The failure mode this prevents is **confident-but-unchecked**: a step returns pl---+## Daily Journal (ALWAYS APPLY)++The daily journal is an **ambient** behavior, not a command. The trigger lives here because it has to be loaded in every session; the procedure lives in the skill (`.claude/skills/daily-journal/SKILL.md`).++- **After finishing a meaningful unit of work, append one entry** to `01-daily/journal/YYYY-MM-DD.md` (get the date with `date +%F`, never guess). Meaningful = shipped or committed something, produced a deliverable, made a decision, changed direction, or hit a notable blocker.+- **Create the file from the skill's template on the first entry of the day.** Append only; newest entries at the bottom of the `## Log` section.+- **Do NOT log** trivial reads, one-line lookups, mid-task scratch work, the journal's own writes, or anything you asked to keep out. Do not announce the write — append and carry on.+- Read the skill body for the entry format, `reflect` mode, and the full guardrails before the session's first write.+- If you say to stop journaling, stop for the rest of the session and do not re-ask.++---+## Integration PreferencesBefore using any external integration in a skill, check `00-inbox/MY-INTEGRATIONS.md`:diff --git a/COG-VERSION b/COG-VERSIONindex 30291cb..f870be2 100644--- a/COG-VERSION+++ b/COG-VERSION@@ -1 +1 @@-3.10.0+3.10.1diff --git a/marketplace-entry.json b/marketplace-entry.jsonindex 9cbce94..4bed0be 100644--- a/marketplace-entry.json+++ b/marketplace-entry.json@@ -7,7 +7,7 @@"url": "https://github.com/huytieu"},"description": "Self-evolving second brain with 33 AI skills, 10 agents, and people CRM. Closed-loop harness release: V-model verification lifecycle with criterion-traced evidence, read-only verifier agents, risk lanes, multi-session ultragoals with a north-star acceptance gate, craft skills for writing and visuals, and paired anti-slop design skills for marketing and product UI. Inspired by Garry Tan's gstack and gbrain, and dwarves-kit. Works with Claude Code, Antigravity, Cursor, Kiro, Gemini CLI, Codex.",- "version": "3.10.0",+ "version": "3.10.1","keywords": ["second-brain","garry-tan",diff --git a/plugin.json b/plugin.jsonindex 4f87822..30c75f2 100644--- a/plugin.json+++ b/plugin.json@@ -1,7 +1,7 @@{"$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json","name": "cog-second-brain",- "version": "3.10.0",+ "version": "3.10.1","description": "Self-evolving second brain with 33 AI agent skills, 10 agents, and people CRM. Closed-loop harness: a V-model verification lifecycle (spec to evidence, criterion-traced checkpoints), read-only verifier agents that observe artifacts instead of trusting worker summaries, risk lanes, multi-session ultragoals with a north-star acceptance gate, plus craft skills for writing, visual work, and a paired anti-slop design system for marketing and product UI. Works with Claude Code, Antigravity, Cursor, Kiro, Gemini CLI, Codex, and any Agent Plugins conformant client.","author": {"name": "Huy Tieu",diff --git a/skills/daily-journal/SKILL.md b/skills/daily-journal/SKILL.mdindex 6f895ca..a8913b4 100644--- a/skills/daily-journal/SKILL.md+++ b/skills/daily-journal/SKILL.md@@ -6,7 +6,7 @@ description: A passive daily work journal that Claude keeps FOR you so you never# Daily Journal## Purpose-Keep an **automatic, passive** journal of your workdays so you don't have to write one. Claude logs what got done as it happens; you optionally run a reflection later. This is distinct from `/daily-checkin` (which is a manual reflection where you supply the input) — here **Claude is the author** and the log accrues in the background.+Keep an **automatic, passive** journal of your workdays so you don't have to write one. Claude logs what got done as it happens; you optionally run a reflection later. This is distinct from `/weekly-checkin` (which is a manual reflection where you supply the input) — here **Claude is the author** and the log accrues in the background.Two modes:1. **`log`** (default, mostly invoked implicitly) — append a short entry to today's journal.@@ -17,7 +17,9 @@ Two modes:- Create the file from the template below on the first entry of the day.- Append only — never rewrite earlier entries. Newest entries go at the bottom of the Log section.-## When to log (behavioral trigger — applies in ANY session, not just when invoked)+## When to log (procedure for the always-apply trigger)+The trigger itself lives in `CLAUDE.md` § Daily Journal (ALWAYS APPLY), because a skill body is lazily loaded and cannot fire itself. This section is the procedure that trigger runs.+After finishing a **meaningful** unit of work, append one entry. Meaningful = would matter to future-you or shows how the day was spent. Examples:- Shipped/committed something, published a note/brief/slide, filed or moved tracker issues (Linear/Jira/GitHub).- Made a decision, changed direction, or hit a notable blocker.