update
Skill3 starsChanged 7 months agoMITView on GitHub
Update mister-anderson plugin and clean up legacy local copies that are now provided by the plugin system.
- Deletes or force-pushes
61 other projects use a version of this.
Adopted by
Most recently taken up by
- orgiggly/claude-plugintheir own version, first seen Nov 8, 2026
- ljchang/mechatheir own version, first seen Nov 8, 2026
- onetapstudiogames/1f3d9-citylifetheir own version, first seen Nov 8, 2026
- patrickjaja/claude-desktop-extratheir own version, first seen Nov 6, 2026
- egregore-labs/egregoretheir own version, first seen Nov 6, 2026
- buildomator/buildomatortheir own version, first seen Nov 4, 2026
Used with
Projects with it also use end, create 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.
torarnvSkillend
Wrap up a work session. Updates memory files, logs decisions and lessons, appends to the session log.
Also in 3 projects that use this
MicrosoftSkillcreate
Also in 4 projects that use this
bitwize-music-studioSkilltutorial
Provides interactive guided album creation for new users. Use when the user is new to the plugin or asks for a walkthrough of the album creation process.
Also in 3 projects that use this
MicrosoftSkillnote
Use this skill to record progress, partial findings, hypotheses, and page-anchored evidence while investigating a long document. Call Note aggressively — any time you read a page and confirm, narrow, or update your understanding of the user's question. Triggers include "jot that down", "save this finding", "note that page 7 says X", or (more often) implicit internal signals — you just learned something and want it recoverable later. Notes persist in the session and Review can select among them. Never paraphrase notes back from memory; write them as you go.
Also in 4 projects that use this
kv0906Skillinbox
Quick capture mode (with content) or batch processing mode (no content). GTD-style routing of inbox items to proper folders. Use "/inbox thought to capture" or "/inbox" to process.
Also in 4 projects that use this
jholhewresSkillbackup
Backup and restore — files, databases, and system backups
Also in 3 projects that use this
- +2 more from microsoft
Other versions61 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 61
- anthropics/knowledge-work-plugins★ 35ksame name
- Q00/ouroboros★ 6.3ksame name
- vellum-ai/vellum-assistant★ 1.5ksame name
- patrickjaja/claude-desktop-extra★ 734same name
- me2resh/apexyard★ 542same name
- daloopa/investing★ 488same name
- TokenRollAI/cc-plugin★ 483same name
- egregore-labs/egregore★ 293same name
- k1p1l0/claude-telegram-supercharged★ 144same name
- kv0906/pm-kit★ 138same name
51 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.
In this repository21 other files it gives its agents.
- skills/setup/templates/AGENTS.mdAGENTS.md
- skills/setup/templates/CLAUDE.mdCLAUDE.md
- add-supervisorSkill
- architectureSkill
- core-loopSkill
- doSkill
- implementationSkill
- investigateSkill
- manifestoSkill
- planSkill
- productSkill
- qualitySkill
- requirementsSkill
- researchSkill
- reviewSkill
- setupSkill
- specificationSkill
- specSkill
- subagents-disciplineSkill
- tasksSkill
- workflowSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 59 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
Update the mister-anderson plugin and clean up legacy local copies from older versions.
- Distinctive: line 32
If remote fetch fails, fall back to reading the local plugin cache:
- Very specific: line 141
Read the file. Locate the YAML frontmatter (between the opening --- and closing ---).
- Very specific: line 160
Read the file. Locate the <on-completion> section (inside the embedded beads-workflow block).
- Distinctive: line 220
If any supervisor migration was skipped, list it explicitly so the user knows enforcement is not active for that supervisor.
“Common” is measured: kept by at least 6 in 10 of the 59 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
Update
Update the mister-anderson plugin and clean up legacy local copies from older versions.
<on-init>
This skill has no arguments. It checks versions and detects legacy files automatically.
</on-init><on-check>
Compare installed version with the latest remote version.LOCAL_VERSION=$(cat .claude/.mister-anderson-version 2>/dev/null | tr -d '[:space:]')
echo "Installed version: ${LOCAL_VERSION:-unknown}"REMOTE_VERSION=$(curl -sf --max-time 5 \
https://raw.githubusercontent.com/websublime/mister-anderson/main/.claude-plugin/plugin.json \
2>/dev/null | grep '"version"' | head -1 | sed 's/[^0-9.]//g')
echo "Remote version: ${REMOTE_VERSION:-fetch failed}"If remote fetch fails, fall back to reading the local plugin cache:
cat ${CLAUDE_PLUGIN_ROOT}/.claude-plugin/plugin.json</on-check><on-check-fail if="version_unknown">
This project may not have been set up with `/setup`. Ask user if they want to proceed anyway.
</on-check-fail><on-execute>What the Plugin Provides vs What Stays Local
Provided by Plugin (no local copy needed):
- Skills, core agents, hooks
Local to Project (never touched by update):
- CLAUDE.md, AGENTS.md, BEADS-WORKFLOW.md
- Dynamic supervisors (
*-supervisor.mdin.claude/agents/) .beads/database, project settings, version file
Step 1: Detect Legacy Local Copies
LEGACY_SKILLS=$(ls -d .claude/skills/*/ 2>/dev/null | wc -l | tr -d '[:space:]')
LEGACY_HOOKS=$(ls .claude/hooks/ 2>/dev/null | wc -l | tr -d '[:space:]')
LEGACY_AGENTS=0
CORE_AGENTS=(architect product-manager research discovery code-reviewer qa-gate beads-owner)
for agent in "${CORE_AGENTS[@]}"; do
[[ -f ".claude/agents/${agent}.md" ]] && LEGACY_AGENTS=$((LEGACY_AGENTS + 1))
done
echo "Legacy skills: $LEGACY_SKILLS"
echo "Legacy hooks: $LEGACY_HOOKS"
echo "Legacy core agents: $LEGACY_AGENTS"List dynamic supervisors that will be PRESERVED:
ls .claude/agents/*-supervisor.md 2>/dev/null | grep -v "^$"If no legacy files found: inform user the project is clean, skip to Step 3.
If legacy files found: present summary and ask user for confirmation.
Step 2: Cleanup Legacy Local Copies
After user confirms:
# Remove legacy local skills
rm -rf .claude/skills/
# Remove legacy local hooks
rm -rf .claude/hooks/
# Remove legacy core agents ONLY (preserve dynamic supervisors)
CORE_AGENTS=(architect product-manager research discovery code-reviewer qa-gate beads-owner)
for agent in "${CORE_AGENTS[@]}"; do
rm -f ".claude/agents/${agent}.md"
done
rm -f ".claude/agents/refactoring-supervisor.md"NEVER remove *-supervisor.md files created by Discovery.
Update version file:
echo "{NEW_VERSION}" > ./.claude/.mister-anderson-versionStep 3: Migrate Supervisors for New Plugin Features
Newer plugin versions may introduce frontmatter or workflow contracts that existing supervisors don't have. Supervisors are local to the project (never removed by cleanup) so they need in-place migration.
Current migrations (as of plugin version introducing state-dimension enforcement):
hooks:block missing from supervisor frontmatter — required forSubagentStopenforcement.bd set-state impl=donestep missing from embedded<on-completion>— required so the agent closes the enforcement state before the hook checks it.
Detect supervisors needing migration
SUPERVISORS=$(ls .claude/agents/*-supervisor.md 2>/dev/null)
[[ -z "$SUPERVISORS" ]] && echo "No supervisors to migrate" || {
echo "Supervisors found:"
for f in $SUPERVISORS; do
NEEDS_HOOKS=$(grep -q "verify-state.sh" "$f" && echo "no" || echo "yes")
NEEDS_STATE=$(grep -q "set-state.*impl=done" "$f" && echo "no" || echo "yes")
echo " $f — hooks: $NEEDS_HOOKS, set-state: $NEEDS_STATE"
done
}Patch each supervisor that needs migration
Do not attempt this with sed/awk. Per-supervisor layout varies (different tech stacks, different specialty content injected by discovery). Use Read + Edit tools so you can see the exact structure before modifying.
For each supervisor file where NEEDS_HOOKS=yes or NEEDS_STATE=yes:
(a) Frontmatter hooks: block — if NEEDS_HOOKS=yes:
Read the file. Locate the YAML frontmatter (between the opening --- and closing ---). Immediately before the closing ---, insert:
hooks:
PreToolUse:
- matcher: Bash
hooks:
- type: command
command: ${CLAUDE_PLUGIN_ROOT}/hooks/stamp-pending.sh
Stop:
- hooks:
- type: command
command: ${CLAUDE_PLUGIN_ROOT}/hooks/verify-state.shPreserve all existing frontmatter fields (name, description, model, tools, etc.) exactly.
(b) Embedded <on-completion> set-state step — if NEEDS_STATE=yes:
Read the file. Locate the <on-completion> section (inside the embedded beads-workflow block). Find the step that calls bd comments add {BEAD_ID} "COMPLETED: and ends with ". Immediately after that step's code block closes, insert a new numbered step:
3. **Record implementation state (MANDATORY — enforced by SubagentStop hook):**
```bash
bd set-state {BEAD_ID} impl=done --reason "Implementation completed on branch {branch-name}"
Renumber subsequent steps accordingly (Push to remote → 4, Clean up labels → 5, etc.). The numbering must be sequential because the workflow is read step-by-step.
#### Post-patch verification
Re-run the detection block. Every supervisor must now show `hooks: no, set-state: no` (both patches applied).
**If any supervisor still fails**, do NOT force-push a broken state. Report the file and ask the user whether to:
- Retry the patch (may have had a parse issue)
- Regenerate that supervisor via `/add-supervisor <tech>` (Step 5 flow)
- Skip (user will patch manually)
---
### Step 4: Post-Update Verification
```bash
ls .claude/agents/*-supervisor.md 2>/dev/null | grep -v "^$"
ls -d .claude/skills/*/ 2>/dev/null && echo "WARNING: Local skills still exist" || echo "OK: No local skills"
ls .claude/hooks/ 2>/dev/null && echo "WARNING: Local hooks still exist" || echo "OK: No local hooks"Step 5: Optional — Re-run Discovery
Offer to refresh dynamic supervisors. If user agrees:
Agent(
subagent_type="discovery",
prompt="Detect tech stack and create/update supervisors for this project. Do NOT delete existing supervisors. Write supervisor files to .claude/agents/."
)Do NOT add extra parameters unless the user explicitly requests it.
</on-execute><on-complete>
Report result:
```
mister-anderson updated to {NEW_VERSION}Cleaned up: {N} local skills, {N} local hooks, {N} core agents Migrated: {N} supervisors patched (hooks block / impl=done step) Preserved: {N} dynamic supervisors (specialty content intact), CLAUDE.md, AGENTS.md
Skills, core agents, and hooks are now provided by the plugin system. Dynamic supervisors remain in .claude/agents/ with enforcement hooks attached.
If any supervisor migration was skipped, list it explicitly so the user knows enforcement is not active for that supervisor.
</on-complete>
<on-next>
**Important:** The version file has been updated, but Claude Code's plugin cache may still hold the previous version.
1. Open the plugin menu: type `/plugin` or go to **Plugins → Installed**
2. Select **mister-anderson** and click **"Update now"**
3. **Restart Claude Code** for the new skills, agents, and hooks to take effect
</on-next>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.

