Agent skill · Audited from source

Update Cog

skill · huytieu/COG-second-brainMIT licenceAudited 14 Oct 2026 · source 6ea1a6ff2160

★ 1,281pushed 11 days ago
✓SAFEGrade B · 89/100 · 14 Oct 2026 · 6ea1a6f
1,281repo stars · whole huytieu/COG-second-brain

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 6ea1a6ff2160 and found no filesystem, network or shell access and no hard-coded secrets. What it does: Check for and apply upstream COG framework updates (skills, docs, scripts) without touching personal content

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Update Cog (huytieu/COG-second-brain)”. Retrieved 14 Oct 2026. https://modelsatlas.com/skills/huytieu/update-cog

Trust audit

Audit 0.4.1 · source read 14 Oct 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium0 low0 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 13,985 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (0)

No findings at any severity. We re-audit when the source changes.

Audit 0.4.1 · source sha 6ea1a6ff2160 · 14 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherhuytieu GitHub userOwner of the repository this skill is read from
  • Source folderhuytieu/COG-second-brain/.claude/skills/update-cogThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdict6ea1a6ff2160A new audit runs when the source changes
  • Repository statusNot archivedLast push 11 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Check for and apply upstream COG framework updates (skills, docs, scripts) without touching personal content

---
name: update-cog
description: Check for and apply upstream COG framework updates (skills, docs, scripts) without touching personal content
roles: [all]
integrations: [git]
---

# COG Update Skill

## Purpose
Help the user update their COG framework files (skills, documentation, scripts) from the official upstream repository without risking their personal content (braindumps, profiles, notes).

## When to Invoke
- User asks to "update COG", "check for updates", or "get latest COG version"
- User mentions wanting new skills or features from upstream
- User asks about their COG version

## Process Flow

### 1. Check Current Version
Read `COG-VERSION` from the vault root. If it doesn't exist, inform the user they may be on an older version that predates version tracking.

### 2. Ensure Upstream Remote
```bash
# Add the upstream remote if not already present
git remote get-url cog-upstream 2>/dev/null || \
  git remote add cog-upstream https://github.com/huytieu/COG-second-brain.git

# Fetch latest
git fetch cog-upstream main --quiet
```

### 3. Compare Versions
```bash
# Local version
cat COG-VERSION

# Upstream version
git show cog-upstream/main:COG-VERSION
```

If versions match, tell the user they're up to date.

### 4. Show What Changed
For each framework file, compare local vs upstream:
```bash
git diff HEAD..cog-upstream/main -- <file>
```

**Framework files** (safe to update — never contain user content):
- Core docs: `README.md`, `SETUP.md`, `AGENTS.md`, `GEMINI.md`, `CHANGELOG.md`, `CONTRIBUTING.md`
- Skills: `.claude/skills/*/SKILL.md`
- Powers: `.kiro/powers/*/POWER.md`
- Gemini: `.gemini/commands/*.toml`, `.gemini/skills/*.md`
- Config: `.claude-plugin/plugin.json`, `marketplace-entry.json`, `.gitignore`
- Scripts: `cog-update.sh`
- Version: `COG-VERSION`

### 5. Detect Customizations
Before updating, check if the user has customized any framework files:
```bash
# Show how the working tree differs from current upstream HEAD.
# NOTE: this cannot tell a local customization apart from an upstream
# change you have not pulled yet — both show up as a diff. Treat any
# output as "needs a look", not as proof the user edited the file.
git diff cog-upstream/main -- <file>
```

If a file differs, warn the user and offer options:
- **Keep yours**: Skip this file
- **Use upstream**: Overwrite with the new version
- **Backup + update**: Save current as `<file>.backup-YYYYMMDD` then update

### 6. Apply Updates
For files the user approves:
```bash
# Surgical file replacement — no merge, no rebase, zero conflict risk
git checkout cog-upstream/main -- <file>
```

### 7. Verify & Summarize
After updating:
- Show the new version number
- List all files that were updated
- List any files the user chose to skip
- Suggest committing the update:
  ```
  git add -A && git commit -m "Update COG framework to v<new-version>"
  ```

## Alternative: Shell Script
For users who prefer a non-AI update, mention the update script:
```bash
./cog-update.sh           # Interactive
./cog-update.sh --check   # Just check for updates
./cog-update.sh --dry-run # Preview changes
./cog-update.sh --force   # Update everything
```

## Important Notes
- **Content folders are NEVER touched**: `00-inbox/`, `01-daily/`, `02-personal/`, `03-professional/`, `04-projects/`, `05-knowledge/`, `06-templates/` contain user data and are always ignored
- **The .gitignore is designed** so content folders are excluded from upstream tracking (only `.gitkeep` files are tracked)
- **The update script updates itself** — `cog-update.sh` is in the framework file list
- **No merge conflicts possible** — this uses `git checkout` for surgical file replacement, not `git merge` or `git rebase`

Excerpt, first 3,704 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 6ea1a6f

Download folder ↗
git clone https://github.com/huytieu/COG-second-brain
git -C COG-second-brain checkout 6ea1a6ff2160

The author documents no install command, so this uses the repository directly. Our verdict covers commit 6ea1a6ff2160 only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source 6ea1a6ff2160

README badge

For maintainers · shows the latest audit

ModelsAtlas: SAFE BMODELSATLASSAFE · B
[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/huytieu/update-cog/badge.svg)](https://modelsatlas.com/skills/huytieu/update-cog/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Similar skills

Add this badge to your README

[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/huytieu/update-cog/badge.svg)](https://modelsatlas.com/skills/huytieu/update-cog/)

Questions about Update Cog

What does the Update Cog skill do?
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.
Is Update Cog safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Update Cog access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (6ea1a6ff2160), read on 2026-10-14. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.