start-session
Skill0 starsChanged 3 months agoMITView on GitHub
Orient a new session in any project directory. Detects empty directory, existing codebase, or resume state, gathers cheap signals, and routes to wayfinder. The only user-invoked entry point.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- abhayla/claude-best-practicestheir own version, first seen Nov 1, 2026
- kartykk/agent-readytheir own version, first seen Sep 21, 2026
- edwardyen724-g/claude-session-skillstheir own version, first seen Aug 30, 2026
- vishalsachdev/claude-skillstheir own version, first seen Aug 12, 2026
- MetaMindTeam/ssot-kittheir own version, first seen Jul 21, 2026
- YehudaFrankel/clankbraintheir own version, first seen Apr 16, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- YehudaFrankel/clankbrain★ 16same name
- abhayla/claude-best-practices★ 9same name
- alxyrgin/agent-forge★ 4same name
- vishalsachdev/claude-skills★ 2same name
- MetaMindTeam/ssot-kit★ 1same name
- edwardyen724-g/claude-session-skills★ 0same name
- kartykk/agent-ready★ 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 repository19 other files it gives its agents.
- codebase-designSkill
- code-reviewSkill
- diagnosing-bugsSkill
- document-directorySkill
- domain-modelingSkill
- grillingSkill
- grill-with-docsSkill
- handoffSkill
- implementSkill
- improve-codebase-architectureSkill
- prototypeSkill
- researchSkill
- resolving-merge-conflictsSkill
- scope-lockSkill
- setupSkill
- tddSkill
- to-specSkill
- to-ticketsSkill
- wayfinderSkill
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 3 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Run once at the beginning of a conversation to orient.
- Very specific: line 27
Check whether .agent/state.json exists in the project root.
- Very specific: line 33
Read .agent/state.json in full.
- Very specific: line 34
Check schema version: If version is 1 (v1 state), tell the user: "This project has v1 state from an earlier version of the plugin.
- Very specific: line 38
Read the wayfinder map (at map_ref path) and any CONTEXT.md or ADRs.
- Very specific: line 72
If size class is medium or large, invoke skills/scope-lock/SKILL.md before any broad reading.
- Distinctive: line 79
Never start implementation before wayfinder has completed its map.
- Very specific: line 80
Never re-run the full orient when .agent/state.json already exists (use Path A instead).
- Distinctive: line 82
Never read broadly across a medium/large repo before scope-lock runs.
“Common” is measured: kept by at least 6 in 10 of the 6 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
Start Session
Run once at the beginning of a conversation to orient. Gathers cheap signals and routes to wayfinder.
Bundled skills (always available)
The following mattpocock/skills ship with this plugin in
skills/ — no external detection or installation needed:
grilling, grill-with-docs, domain-modeling, tdd, prototype,
codebase-design, diagnosing-bugs, improve-codebase-architecture
These are referenced at specific integration points by wayfinder, implement, to-spec, and code-review.
Process
Step 1: Determine which path applies
Check whether .agent/state.json exists in the project root.
Path A: Resume (.agent/state.json exists)
- Read
.agent/state.jsonin full. - Check schema version. If
versionis1(v1 state), tell the user: "This project has v1 state from an earlier version of the plugin. Would you like me to migrate it to v2, or start fresh?" Act on their answer. - Read the wayfinder map (at
map_refpath) and anyCONTEXT.mdor ADRs. - Summarize current status to the user: what the destination is, which tickets are done, which are next.
- Resume: read the map and pick up the next unresolved ticket.
Path B: Empty directory (no substantive files)
Count files excluding .git*, .agent*, README*, LICENSE*,
.gitignore, .gitattributes. If zero:
- Tell the user: "This directory is empty — treating as a new project."
- Invoke
skills/setup/SKILL.mdto configure the tracker. - Invoke
skills/wayfinder/SKILL.md— the first ticket on the map will be a grilling ticket to establish what the project is.
Path C: Existing codebase (has files)
Gather cheap heuristic signals:
- Tracked files —
git ls-files | wc -l(or directory file count). - Approximate LOC — count lines across common source extensions.
- Existing documentation — check for
CONTEXT.md,docs/CONTEXT.md,docs/agents/,docs/adr/,docs/architecture/. - Size class —
large(300+ files or 30k+ LOC),medium(40+ or 3k+),small(everything else).
Report signals to the user, then:
- Invoke
skills/setup/SKILL.mdto configure the tracker. - If size class is
mediumorlarge, invokeskills/scope-lock/SKILL.mdbefore any broad reading. - Invoke
skills/wayfinder/SKILL.md— the first ticket on the map will be a grilling ticket to establish what the user wants to do.
What this skill must never do
- Never start implementation before wayfinder has completed its map.
- Never re-run the full orient when
.agent/state.jsonalready exists (use Path A instead). - Never read broadly across a medium/large repo before scope-lock runs.
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.

