| name | cc-best-practice-adoption |
| description | Use when onboarding new Codex features, reviewing external best-practice repos, or expanding the daily learning catalog. Covers: gap analysis against native CC capabilities, tips catalog expansion, daily-learning script, native .Codex/ structure (agents, commands, settings). |
| version | 1.0.0 |
| updated | "2026-04-03T00:00:00.000Z" |
| category | workspace-hub |
| type | skill |
| tags | ["learning","onboarding","Codex","best-practice","daily-cadence"] |
| related_skills | ["comprehensive-learning"] |
| platforms | ["linux"] |
CC Best Practice Adoption
Workflow for analyzing external Codex best-practice sources, distilling actionable
gaps in our ecosystem, and wiring daily learning into the session lifecycle.
When to Use
- New Codex version ships with features to evaluate
- Reviewing an external repo (e.g., shanraisshan/Codex-best-practice)
- Expanding the daily learning tip catalog
- Bootstrapping native .Codex/ structure (agents, commands, skills)
Gotchas
1. Native vs Custom — Check Before Building
Critical lesson learned: /powerup, /insights, /stats, /context, /simplify,
/batch, /debug, /diff, /security-review are all native CC commands (v2.1.90+).
We wasted effort planning to build custom versions before discovering they exist.
Rule: Before creating any slash command, run /help or check
Codex-best-practice/best-practice/Codex-commands.md
for the 64 native commands.
2. Hermes Skills vs Native CC Skills
Our ecosystem has 691 Hermes-style skills but ZERO native CC format. They are different:
- Native CC skills:
.Codex/skills/<name>/SKILL.md with CC frontmatter (context: fork, paths, model, effort, hooks)
- Hermes skills:
~/.hermes/skills/<name>/SKILL.md with Hermes frontmatter
- Both can coexist. Native CC skills show in
/skills menu and support auto-discovery.
3. Tips Catalog YAML Is Not Standard YAML
config/workflow-tips/tips-catalog.yaml uses a minimal structure parsed by
daily-learning.py without PyYAML. Don't add complex YAML features (anchors, flow
sequences in values). Keep entries as simple key: value pairs.
4. Deterministic Daily Seed
daily-learning.py uses md5(YYYY-MM-DD) as seed so the same 2 tips show all day.
Running it multiple times won't cycle tips — that's intentional (spaced repetition).
Workflow Steps
