session-startup
Ensures the agent is oriented on the current project state before taking action. Prevents the agent from working with stale context or making assumptions about project status. [GENERATED BY MOTIF — review and customize]
Best use case
session-startup is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Ensures the agent is oriented on the current project state before taking action. Prevents the agent from working with stale context or making assumptions about project status. [GENERATED BY MOTIF — review and customize]
Teams using session-startup should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/session-startup/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How session-startup Compares
| Feature / Agent | session-startup | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Ensures the agent is oriented on the current project state before taking action. Prevents the agent from working with stale context or making assumptions about project status. [GENERATED BY MOTIF — review and customize]
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Session Startup Ensures the agent is oriented on the current project state before taking action. Prevents the agent from working with stale context or making assumptions about project status. ## When to Use - User opens a new session with any variant of 'refresh yourself' - User says 'we're working in [project] today' or 'we shall continue' - Especially when: multiple days have passed since last session - Don't skip when: returning to a project after working on a different one ## Instructions 1. Read project README and any STATUS.md or MEMORY.md files 2. Check recent git log (last 5-10 commits) for what changed 3. Summarize: current phase, what's working, what's blocked, what's next 4. If deployment state matters, check production status 5. Wait for user direction before acting - do not propose work unprompted 6. If user provides specific direction, acknowledge context and proceed ## Best Practices - Keep the summary concise (3-5 bullet points) - Mention specific recent changes by name - **Avoid:** Don't dump entire file contents -- summarize - **Avoid:** Avoid proposing work until asked ## Common Pitfalls | Problem | Solution | |---|---| | Agent starts working immediately without reading project state | Always read STATUS/README first, even if you think you remember | | Summary is too long and buries the key points | Lead with phase and blockers, keep to 3-5 bullets | ## Key Constraints - Never propose work before the user gives direction - Always read project status files before summarizing - Summary must be concise enough to read in 10 seconds