Skip to content

softaworks/agent-toolkit / skills/session-handoff

session-handoff

Verified

Creates comprehensive handoff documents for seamless AI agent session transfers. Triggered when: (1) user requests handoff/memory/context save, (2) context window approaches capacity, (3) major task milestone completed, (4) work session ending, (5) user says 'save state', 'create handoff', 'I need to pause', 'context is getting full', (6) resuming work with 'load handoff', 'resume from', 'continue where we left off'. Proactively suggests handoffs after substantial work (multiple file edits, complex debugging, architecture decisions). Solves long-running agent context exhaustion by enabling fresh agents to continue with zero ambiguity.

FileKindSize
README.mdmarkdown6.9 KB
SKILL.mdskill md6.2 KB
evals/model-expectations.mdmarkdown5.6 KB
evals/results-opus-baseline.mdmarkdown2.6 KB
evals/setup_test_env.pyexecutablescript10.1 KB
evals/test-scenarios.mdmarkdown5.2 KB
references/handoff-template.mdmarkdown3.9 KB
references/resume-checklist.mdmarkdown2.7 KB
scripts/check_staleness.pyexecutablescript12.7 KB
scripts/create_handoff.pyexecutablescript10.4 KB
scripts/list_handoffs.pyexecutablescript3.7 KB
scripts/validate_handoff.pyexecutablescript10.6 KB

references/resume-checklist.md

View on GitHub
# Resume Checklist

Follow this checklist when resuming work from a handoff document to ensure zero-ambiguity continuation.

## Pre-Resume Verification

- [ ] Read the entire handoff document before taking any action
- [ ] Verify you are in the correct project directory
- [ ] Confirm the git branch matches (or understand why it might differ)
- [ ] Check the handoff timestamp - how stale is this context?

## Context Validation

- [ ] Review "Important Context" section thoroughly
- [ ] Understand all assumptions listed - are they still valid?
- [ ] Check if any blockers have been resolved since handoff
- [ ] Review "Potential Gotchas" to avoid known pitfalls

## State Verification

- [ ] Run `git status` to see current file state
- [ ] Compare modified files list in handoff vs current state
- [ ] Check if any environment variables need to be set
- [ ] Verify any required services/processes are running

## Resume Execution

- [ ] Start with "Immediate Next Steps" item #1
- [ ] Reference "Files Modified" table for context on recent changes
- [ ] Apply patterns documented in "Key Patterns Discovered"
- [ ] Follow architectural insights from "Architecture Overview"

## During Work

- [ ] Update handoff document if major new context is discovered
- [ ] Mark completed items in "Pending Work" as you finish them
- [ ] Add new blockers/questions as they arise
- [ ] Consider creating a new handoff if session becomes long

## Red Flags - Stop and Verify

If you encounter any of these, pause and verify context before proceeding:

1. **Files mentioned in handoff don't exist** - codebase may have changed significantly
2. **Branch has diverged substantially** - check git log for recent commits
3. **Assumptions are clearly invalid** - reassess the approach
4. **Blockers marked as unresolved are now blocking you** - escalate to user
5. **Architecture has changed** - re-explore before continuing

## Quick Start Commands

After reading the handoff, these commands help verify state:

```bash
# Check current branch and status
git branch --show-current
git status

# See recent commits (compare with handoff)
git log --oneline -10

# Check for any running processes mentioned
ps aux | grep [process-name]

# Verify environment
env | grep [relevant-var]
```

## Handoff Quality Assessment

Rate the handoff quality to identify if more exploration is needed:

| Aspect | Good | Needs Exploration |
|--------|------|-------------------|
| Next steps | Clear, actionable | Vague or missing |
| File references | Specific paths/lines | General descriptions |
| Decisions | Rationale included | Just outcomes |
| Context | Complete picture | Gaps or assumptions |

If multiple aspects "Need Exploration", spend time re-exploring the codebase before continuing implementation.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall