update-backlog
Skill1 starsChanged 7 months agoMITView on GitHub
Update BACKLOG.md at the end of a working session. Trigger this skill when the user says "update the backlog", "end of session", "wrap up", or "what did we do". Also run it automatically after completing any backlog task, even if the user doesn't ask — it should never be skipped.
In this repository12 other files it gives its agents.
- api-consistency-checkSkill
- architecture-drift-checkSkill
- design-reviewSkill
- jml-completeness-checkSkill
- jml-conformance-checkSkill
- jml-designSkill
- jml-test-generationSkill
- new-classSkill
- refactor-classSkill
- static-analysisSkill
- test-coverage-reviewSkill
- update-readmeSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
After completing, partially completing, or abandoning a backlog task
- Distinctive: line 16
After discovering a new issue or design decision worth tracking
- Distinctive: line 44
Add a new section if a new class or module was introduced this session.
- Distinctive: line 48
Follow the existing format exactly:
- Very specific: line 58
If a task has a dependency or constraint, append it after — (e.g.
- Distinctive: line 63
Before writing, show the user:
- Distinctive: line 68
Ask for confirmation if anything is ambiguous.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Update Backlog
When to run
- At the end of every working session
- After completing, partially completing, or abandoning a backlog task
- After discovering a new issue or design decision worth tracking
Procedure
1. Identify changes from the session
Review the conversation and extract:
- Tasks that were fully completed → move to
### Doneunder the correct section, using[x] - Tasks that were partially completed → keep under the section with
[ ], add a note like(in progress — [what remains]) - Tasks that were abandoned or put on hold → keep with
[ ], update or add— on holdwith a short reason - New tasks or issues discovered → add under the correct section with
[ ] - New design decisions → these belong in
ARCHITECTURE.md, not the backlog; flag them separately for the user
2. Determine the correct section
Each task belongs to one of:
## Project-wide— affects the whole project or no specific class/module## ClassName— specific to a class or module
Add a new section if a new class or module was introduced this session.
3. Write the update
Follow the existing format exactly:
- [ ] Task description — optional note
- [x] Completed task description- One task per line
- No sub-bullets
- Short, imperative phrasing ("Create", "Fix", "Implement", "Change")
- If a task has a dependency or constraint, append it after
—(e.g.— on hold,— blocked by X)
4. Present the diff to the user
Before writing, show the user:
- What you are marking as done
- What you are adding
- What you are modifying
Ask for confirmation if anything is ambiguous.
Format reminder
The full structure of BACKLOG.md is:
# Backlog — [project name]
Pending tasks and known issues, grouped by class or module. Update this
file at the end of each session.
---
## Project-wide
...
## ClassName
...
---
## Done
### ClassName
- [x] ...Preserve all existing content. Only add, tick, or annotate — never delete.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

