- Home
- Skills
- Leegonzales
- Aiskills
- Context Continuity Code
context-continuity-code_skill
- Python
20
GitHub Stars
3
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill leegonzales/aiskills --skill context-continuity-code- CHANGELOG.md409 B
- README.md1.2 KB
- SKILL.md14.6 KB
Overview
This skill preserves and transfers development context specifically for Claude Code sessions. It captures code edits, git state, running services, environment variables, decisions, and open tasks so work can resume with minimal friction. Use it to hand off or resume engineering work while retaining machine and process state.
How this skill works
It inspects the workspace to identify active files, uncommitted changes, recent commits, and branch information, then records running services, ports, environment variables, and background processes. It summarizes technical decisions, peer review inputs, test status, and explicit next actions in a single structured artifact optimized for development handoffs. The artifact is concise (development-optimized) and designed to be pasted into a new Claude Code session or given to another developer.
When to use it
- Starting a fresh Claude Code session and you need prior context
- Handing off in-progress development to another developer or AI agent
- Resuming work after your session hit token/window limits
- Preparing a compact, actionable snapshot before major refactors or merges
- Before requesting peer review (Codex/Gemini) or after integrating their feedback
Best practices
- Run git status/diff and include file:line references for active files
- List branch, recent commits, staged/unstaged/untracked files explicitly
- Record running services, ports, and any required env vars (redact secrets)
- Capture technical decisions and peer review outcomes with rationale and tradeoffs
- Include explicit next actions, blockers, and tests still needed
Example use cases
- Mid-bugfix transfer: capture failing test, stack trace, modified files, and reproduction steps
- Feature continuation: hand off a partially implemented feature with branch, uncommitted changes, and services to run
- Peer review prep: send Code+Git+Technical Decisions to Codex/Gemini for focused feedback
- Disaster recovery: snapshot environment and git state before major refactor or risky merge
- Context window overflow: compress a long session into a transfer artifact for a new Claude Code session
FAQ
No. Redact API keys and credentials and avoid pasting entire files; reference paths and line ranges instead.
What minimal git info is required?
At minimum include current branch, base branch, recent commits, and lists of staged/unstaged/untracked files so the receiver can reproduce the exact repo state.