progress-updating
Skill0 starsChanged 7 months agoMITView on GitHub
Updates PROGRESS.md with session entry and captures exactly where we stopped. Triggers on /progress, update progress, or at end of session before commit. Tracks what was done, what was learned, where we are in the flow, and what comes next. The repo never lies.
In this repository21 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adr-captureSkill
- adversarial-reviewSkill
- architect-firstSkill
- blog-draftingSkill
- brainstorming-knowledge-builderSkill
- code-reviewingSkill
- session-startSkill
- debug-systemsSkill
- evolution-logSkill
- faang-loop-simulationSkill
- learn-in-publicSkill
- mock-assessingSkill
- os-lensSkill
- partner-systemSkill
- review-systemsSkill
- spaced-reviewingSkill
- system-checkingSkill
- systems-traceSkill
- trend-scoutingSkill
- weekly-reflectingSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 47
Every progress entry must tell the next session exactly where to pick up:
- Distinctive: line 54
Every session ends with a progress update + commit.
- Distinctive: line 56
If the student can't articulate a key learning, the session wasn't focused enough.
- Distinctive: line 58
If wiki pages were created during the session, note them in the entry.
≈ “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
Progress Update -- Session Log
The repo never lies. This keeps the record honest, current, and useful for the next session start.
Trigger
/progressor "update progress"- At end of session, before commit
Execution
- Read last entry in PROGRESS.md (get current session number and state).
- Gather from session context (or ask the student if unclear):
- What did we do this session? (specific activities: theory, build, review, brainstorm)
- What's the key thing learned? (one sentence, specific)
- Where exactly are we stopping? (mid-prep? exercise done? review pending?)
- Approximately how long was this session? (even rough: 15m, 1h, 3h)
- What concepts go into the spaced review deck?
Append new row to PROGRESS.md table:
| {session} | {date} | {time} | {topic} | {key learning} | {state} |Time format:
~15m,~1h,~3h. Even rough estimates help future students plan.Session state examples:
exercise complete, reviewed-- fully donetheory discussed, exercise not started-- mid-flowexercise built, review pending-- will review next sessionbrainstorm only -- wiki page created-- knowledge session
- Update ROADMAP.md current status if topic or project state changed.
- If concepts were learned, prompt to add them to spaced review deck.
- Remind: "Run
/postfor your X post, then commit and push."
What Gets Captured
Every progress entry must tell the next session exactly where to pick up:
- What was DONE (so we don't repeat it)
- What is IN PROGRESS (so we continue it)
- What is NEXT (so we can plan the session)
Rules
- Every session ends with a progress update + commit. Non-negotiable.
- Key learning must be specific, not vague. "Learned about dicts" → rejected. "Learned that defaultdict avoids KeyError checks in frequency counting" → accepted.
- If the student can't articulate a key learning, the session wasn't focused enough. Flag it.
- Session state must be precise enough that next session's /checkin can pick up exactly where we left off.
- If wiki pages were created during the session, note them in the entry.
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.

