start
Skill8 starsChanged 5 months agoMITView on GitHub
Session start ritual. Loads context, checks the last closing report, and proposes priorities. Use when the user begins a work session — says "start", "let's begin", "good morning", or asks what to work on today.
139 other projects use a version of this.
Adopted by
Most recently taken up by
- Robo-Co-op/robobuilder-standardtheir own version, first seen Nov 16, 2026
- schen22/claude-configtheir own version, first seen Nov 16, 2026
- futorio/red-greentheir own version, first seen Nov 16, 2026
- tatashbar/Economic-Modeling-Studiotheir own version, first seen Nov 16, 2026
- flavien-ia/hypervibe-harnesstheir own version, first seen Nov 15, 2026
- srinator22/agent-kerneltheir own version, first seen Nov 15, 2026
Used with
Projects with it also use project-stage-detect, team-level and 10 more.
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.
CoralGameSkillproject-stage-detect
Automatically analyze project state, detect stage, identify gaps, and recommend next steps based on existing artifacts.
Also in 18 projects that use this
CoralGameSkillteam-level
Orchestrate level design team: level-designer + narrative-director + world-builder + art-director + systems-designer + qa-tester for complete area/level creation.
Also in 13 projects that use this
lilangMaxSkillcreate-stories
Break a single epic into implementable story files. Reads the epic, its GDD, governing ADRs, and control manifest. Each story embeds its GDD requirement TR-ID, ADR guidance, acceptance criteria, story type, and test evidence path. Run after /create-epics for each epic.
Also in 12 projects that use this
lilangMaxSkillstory-done
End-of-story completion review. Reads the story file, verifies each acceptance criterion against the implementation, checks for GDD/ADR deviations, prompts code review, updates story status to Complete, and surfaces the next ready story from the sprint.
Also in 12 projects that use this
- +3 more from CoralGame
- +5 more from lilangMax
Other versions139 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 139
- Donchitos/Claude-Code-Game-Studios★ 39ksame name
- anthropics/knowledge-work-plugins★ 36ksame name
- openai/plugins★ 7.8ksame name
- SterlingChin/marvin-template★ 1ksame name
- educlopez/ui-craft★ 441same name
- ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tool★ 432same name
- dsifry/metaswarm★ 432same name
- pixel-cellar/Claude-Code-Game-Studios★ 324same name
- ansvisor/ansvisor★ 241same name
- alexeykrol/claude-code-starter★ 194same name
129 more not listed
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.
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. 0 are common, 5 are distinctive, and 9 are very specific to this project.
0
Common
Most of the 126 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Read brain/team.md and count the entries under users:.
- Distinctive: line 14
0 or 1 user (SOLO): run the Solo protocol below.
- Distinctive: line 15
2 or more users (TEAM): run the Team protocol below.
- Very specific: line 17
If brain/team.md does not exist, treat it as SOLO.
- Very specific: line 24
Read brain/context.md (the index — always loaded).
- Very specific: line 25
Read the latest notes/daily-summaries/closing-*.md file (if any).
- Distinctive: line 36
Wait for the response before proceeding.
- Distinctive: line 58
Only the session reports are per-person, so two people can work and close on the same day without overwriting each other.
- Very specific: line 62
Ask: "Who is starting this session?" and match the answer to a handle in brain/team.md.
- Very specific: line 72
Read brain/context.md (the shared index — always loaded).
- Very specific: line 73
If state_isolation: per-user, also read brain/context-<user>.md if it exists (this person's own state.
- Very specific: line 79
Read the latest notes/daily-summaries/<user>/closing-*.md (this person's own last close).
- Very specific: line 80
Read the most recent dated entries in brain/changelog.md (if it exists) to see what teammates changed in shared brain files since this user …
- Distinctive: line 95
If nothing is relevant, say nothing — no filler lines.
“Common” is measured: kept by at least 6 in 10 of the 126 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
Session Start
Execute the session start protocol.
Step 0: Solo or team?
Read brain/team.md and count the entries under users:.
- 0 or 1 user (SOLO): run the Solo protocol below. Do not ask who is here.
- 2 or more users (TEAM): run the Team protocol below.
If brain/team.md does not exist, treat it as SOLO.
Solo protocol
Step 1: Load context
- Read
brain/context.md(the index — always loaded). - Read the latest
notes/daily-summaries/closing-*.mdfile (if any).
Step 2: Propose
Summarize in max 5 bullet points:
- What was done last session (from the closing report).
- What is pending.
- Current focus from the context index.
Then ask: "What are we working on today?"
Step 3: STOP
Wait for the response before proceeding.
Step 4: Load relevant context
Based on the answer, load ONLY the matching file from brain/contexts/:
| User says... | Load |
| Work, clients, sales, deals | brain/contexts/work.md |
| Building, coding, projects | brain/contexts/projects.md |
| Content, blog, social, video | brain/contexts/content.md |
| Anything else | Ask: "Is this work, building, or content?" |
Never load all context files at once. One area per session.
Step 5: Work
Proceed with the task — the right context is loaded.
Team protocol
The brain (brain/context.md, brain/contexts/*, agents, hooks) is shared.
Only the session reports are per-person, so two people can work and close on
the same day without overwriting each other.
Step 1: Identify the user
Ask: "Who is starting this session?" and match the answer to a handle in
brain/team.md.
Skip the question if the user already named themselves in this message, or the
context makes it obvious. If the answer matches no handle, stop and ask again —
never guess. Below, <user> is the matched handle.
Step 2: Load context
First check state_isolation in brain/team.md (absent means shared).
- Read
brain/context.md(the shared index — always loaded).- If
state_isolation: per-user, also readbrain/context-<user>.mdif it exists (this person's own state; on a first session it may not yet — that is fine,/closecreates it). In that modecontext.mdholds only the shared header (team, this-week priorities, cross-person handoffs); each operator's deals and areas live in their own file. Do not read the other operators'context-<handle>.mdfiles.
- If
- Read the latest
notes/daily-summaries/<user>/closing-*.md(this person's own last close). - Read the most recent dated entries in
brain/changelog.md(if it exists) to see what teammates changed in shared brain files since this user last worked.
Step 2.5: Surface only relevant teammate activity
Scan the latest closing report of each other user
(notes/daily-summaries/<other>/closing-*.md). Surface an item ONLY if it is
one of these — otherwise skip it:
- Blocker: a teammate is waiting on
<user>, or<user>is blocked waiting on them. Flag as a P0 callout. - Area overlap: the teammate touched something in
<user>'s area (same client, deal, project, or topic). - Cross-cutting decision: a teammate made a call that changes how
<user>should approach today's work.
Skip routine work that does not affect <user>: internal refactors, doc
maintenance, hooks/scripts, brain housekeeping, and work in areas <user> does
not own. If a teammate's last close is older than 3 days, skip them entirely.
If nothing is relevant, say nothing — no filler lines.
Step 3: Propose
Summarize in max 5 bullet points:
- What
<user>did last session (from their closing report). - What is pending on
<user>'s plate. - Current focus from the context index, filtered to
<user>'s areas.
Then ask: "What are we working on today?"
Step 4: STOP
Wait for the response before proceeding.
Step 5: Load relevant context
Same routing table as the Solo protocol (Step 4). Load ONLY the one matching
file from brain/contexts/. Never load all context files at once.
Step 6: Work
Proceed with the task.
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.

