Install
$ agentstack add skill-heffrey-claude-skills-end-session ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
End Session Workflow
Flexible. Step 3 always applies; steps 1, 2, and 4 are conditional — skip them if the session didn't produce anything worth capturing, or if the target file doesn't exist and creating one isn't warranted. Don't manufacture content to fill a section — say "nothing to add" and move on.
When to use
- Wrapping up a coding session.
- About to step away from a task for a day or longer.
- After finishing a feature, fix, or investigation worth documenting.
Step 1: Update CLAUDE.md with Session Learnings
Review what changed in the codebase during this session and add any new architectural insights, gotchas, or patterns to CLAUDE.md.
What to capture:
- New subsystems or modules added
- Changed behavior in existing systems
- Gotchas discovered (non-obvious constraints, performance traps, ordering requirements)
- New helper functions or patterns that future work should know about
- Configuration files or schemas introduced
Examples to look for:
- Code that surprised you or was trickier than expected
- Interdependencies you discovered between systems
- Performance considerations
- Edge cases that came up
- Data structures that need explanation
Step 2: Update README.md with Relevant Documentation
Add or update user-facing or contributor-facing documentation in README.md.
What to capture:
- New user-visible features
- How to run or use the project
- Build/development instructions (if changed)
- Feature overview
- Known limitations or TODOs visible to users
Step 3: Propose Logical TODOs
Based on the session's work and what you learned:
- Incomplete work — features started but not finished
- Discovered issues — bugs found but not fixed, or architectural problems identified
- Natural next steps — obvious follow-up work that builds on what was done
- Technical debt — refactorings that would make future work easier
- Performance — optimizations that should be considered
Format as:
- [ ] Brief description (context about why it matters)
Example:
- [ ] Add retry logic to the upload client (currently fails silently on transient errors)
- [ ] Extract shared validation into a helper (duplicated across three handlers)
- [ ] Add index on the events table's created_at column (queries slow past ~100k rows)
Step 4: Propose Commit and Push
If uncommitted changes exist, offer to commit them with a summary message and push. This step is conditional — skip it if there's nothing to commit, if the user manages commits themselves, or if the changes aren't ready to ship.
Ask: "There are uncommitted changes — want me to commit and push?"
Don't commit silently. Always confirm before acting.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: heffrey
- Source: heffrey/claude-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.