session-audit
Skill0 starsChanged 6 months agoMITView on GitHub
Simple end-of-session audit that scans for uncaptured corrections, preferences, and decisions, then proposes saving them to the right workspace file. Use this skill whenever you say 'audit this session,' 'session audit,' 'what did we miss,' or 'end of session check.' Works with any Cowork workspace that has a CLAUDE.md and MEMORY.md in the root folder.
9 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- 5hirish/ducttheir own version, first seen Nov 8, 2026
- ethosagent/ethostheir own version, first seen Nov 3, 2026
- zanellig/skillstheir own version, first seen Nov 1, 2026
- nurmdrafi/pi-nativetheir own version, first seen Oct 25, 2026
- iceberggymnast/skillstheir own version, first seen Oct 17, 2026
- GRIDLOCK-NYC/claude-skillstheir own version, first seen Oct 8, 2026
Used with
Projects with it also use handoff.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
MicrosoftSkillhandoff
Compact the current conversation into a handoff document for another agent to pick up.
Also in 3 projects that use this
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- meggiepham06-smile/session-audit-skill★ 021% same
Wrote their own version · 8
- ethosagent/ethos★ 22same name
- 5hirish/duct★ 8same name
- nurmdrafi/pi-native★ 4same name
- t1merickson/skills★ 1same name
- zanellig/skills★ 1same name
- GRIDLOCK-NYC/claude-skills★ 0same name
- iceberggymnast/skills★ 0same name
- Lucioway/way-stack★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository4 other files it gives its agents.
- create-projectSkill
- init-workosSkill
- uninit-workosSkill
- workspace-auditSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
4
Distinctive
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 23
Read these files if they exist:
- Distinctive: line 87
If there are no findings, say so: "Clean session.
- Distinctive: line 91
After you approve (all, some, or none), write the approved changes to the appropriate files.
- Distinctive: line 93
Important: Never write changes without approval.
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Starter Session Audit
A lightweight audit that runs at the end of any Cowork session. It catches things you told Cowork during the session that should be saved permanently so you never have to say them again.
What This Skill Does
Two things, and only two things:
- Scans for uncaptured learnings. Looks through the conversation for corrections you made, preferences you expressed, and decisions you stated that aren't already written in your workspace files.
- Proposes where to save each finding. For each uncaptured learning, it tells you which file it belongs in, what section, and the exact wording. You approve or skip each one.
That's it. No file reorganization, no cleanup, no progress tracking. Just: "Did I learn anything this session that should be remembered?"
Step 1: Discover Your Workspace
Find your workspace root dynamically. Look for a CLAUDE.md file in the mounted workspace folder. Read whatever workspace files exist. The audit adapts to your setup — it works whether you have one project or twenty.
Read these files if they exist:
- Root CLAUDE.md (your standing instructions)
- Root MEMORY.md (your accumulated context)
- Any project CLAUDE.md and MEMORY.md files that were used during this session
- Any reference files that were loaded during this session (e.g., voice-principles.md)
Step 2: Scan the Conversation
Go through the entire conversation from top to bottom. Look for these four signal types:
A. Corrections
You fixed something Cowork produced. Maybe you changed a word, rewrote a sentence, adjusted a format, or said "no, do it this way instead." Each correction reveals a rule that Cowork should follow next time.
What to look for: Moments where you edited, rejected, or rewrote Cowork's output. Ask: what underlying preference or rule drove the change?
Example: You changed "Best regards" to "Thanks" on an email draft. The underlying rule: "Sign off with 'Thanks' for internal contacts."
B. Explicit Preferences
You stated a preference directly. Words like "always," "never," "I prefer," "from now on," "I like it when," or "don't do that."
What to look for: Direct instructions about how you want things done, even casual ones.
Example: "I prefer bullet points over numbered lists." "Don't use exclamation points in subject lines."
C. Decisions
You made a decision that affects future work. Chose one option over another, set a deadline, established a rule for a project, or resolved an ambiguity.
What to look for: Choices that should be recorded so Cowork doesn't re-ask the same question later.
Example: "Let's go with the $5,000 savings target." "Cancel the gym membership, keep Spotify."
D. New Context
You shared a fact about yourself, your work, or your world that Cowork didn't previously know. Contact details, schedules, relationships, project updates, or anything that changes how Cowork should approach future tasks.
What to look for: Information that would help Cowork in a future session if it remembered it.
Example: "My rent went up to $2,600 starting next month." "Sarah is my new manager."
Step 3: Filter Against What's Already Saved
For each finding from Step 2, check whether it's already captured in the workspace files you loaded in Step 1. Skip anything that's already written down. Only surface genuinely new findings.
Step 4: Present Findings
Present each finding in this format:
**[Number]. [What happened]**
- **The rule/fact:** [Exact wording to save, written as a clear instruction or statement]
- **Where it goes:** [File path and section name]
- **Why:** [One sentence on why this matters for future sessions]Group findings into two categories:
Recommend (apply unless you object): Clear-cut findings where the right action is obvious.
Your call: Findings where there's a judgment call or where you might want to phrase the rule differently.
If there are no findings, say so: "Clean session. Nothing new to capture." Don't manufacture findings.
Step 5: Apply Approved Changes
After you approve (all, some, or none), write the approved changes to the appropriate files. For each change, confirm what was written and where.
Important: Never write changes without approval. Always present findings first and wait.
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.

