Agent skill
session-start
Start a session by loading workspace context and prior state
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/session-start-daemn256-workspace-repo-templ-2
SKILL.md
Session Start
Uses Orchestrator. Orchestrator loads context directly — session operations run inline because they need conversation context and judgment. Load workspace context, prior session state, and establish the session goal.
Prerequisites: Access to workspace configuration files
Phase 1: Load Context
Read Workspace Files
Read these files in order:
workspace.config.yaml— process profile, forge topology, board IDs, AI runtimes, commandsdocs/workspace/context.md— tech stack, domain terms, architecture, key conventionsdocs/workspace/goals.md— durable priorities, milestones, backlog (if exists).tmp/workspace/goals.md— active sprint state, current focus (if exists)
Check for Prior Sessions
- List files in
.tmp/sessions/ - Read the most recent session handoff artifact (if any)
- Note any incomplete work or open questions from prior sessions
Verify Board Status
For any tracked issues carried over from prior sessions or identified in sprint state:
- Read
workspace.config.yamlfor board field IDs - Check current board status for each active issue
- Flag any stale statuses (e.g., issue still "In Progress" but branch was merged)
- Correct stale statuses before proceeding
Phase 2: Establish Session
Summarize Context
- State the process profile and its implications
- Note active AI runtimes
- Identify current priorities from goals.md
- Flag any carry-over from prior sessions
State the Goal
- Confirm the session goal with the user
- Identify active constraints
- Acknowledge approval gates
Output
## Context Anchors
- **Issue:** #<number> - <title> (if applicable)
- **Branch:** <active branch> (if applicable)
- **Prior Session:** <session file> (if applicable)
## Session Context
**Process Profile:** <lightweight | standard | regulated>
**AI Runtimes:** <list>
**Active Priorities:**
- <priority 1>
- <priority 2>
## Carry-Over (if any)
- <incomplete item from prior session>
- <open question>
## Session Goal
<Clear statement of what this session aims to accomplish>
## Next Step
<First action toward the goal>
**Approval Required:** Yes
⛔ CHECKPOINT
STOP. Confirm the session goal before proceeding to execution.
Error Handling
| Situation | Recovery |
|---|---|
| workspace.config.yaml missing | Ask user to run setup-workspace |
| goals.md missing or empty | Ask user for current priorities |
| Prior session file references stale state | Verify current state before acting |
| Context files are outdated | Use refresh-context prompt to update |
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?