Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/sfw-reset
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: data
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/reset
What's inside
- Skill name
reset- Description
- Nuclear option — clear all ALM learning data and start fresh. Requires explicit user confirmation.
- License
- MIT
- Instructions
- 45 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# ALM Reset
Delete all ALM learning data and reinitialize to first-run state.
## Instructions
### CRITICAL: Ask for Confirmation First
Before doing anything, ask the user:
> Are you sure you want to reset ALM? This will delete all evaluations, playbooks, confidence scores, and the classifier model. This cannot be undone.
Wait for the user to explicitly confirm (e.g., "yes", "confirm", "do it"). If they say no or anything ambiguous, abort and output: `[ALM] Reset cancelled.`
### Perform Reset
Only after explicit confirmation:
1. Delete all files in `~/.claude/alm/evaluations/` (including the `archive/` subdirectory if it exists)
2. Delete all files in `~/.claude/alm/playbooks/`
3. Delete all files in `~/.claude/alm/classifier/`
4. Delete `~/.claude/alm/confidence.json`
5. Delete `~/.claude/alm/reflect-queue.json`
6. Delete `~/.claude/alm/config.json` (if it exists)
Then reinitialize:
7. Recreate directory structure: `evaluations/`, `playbooks/`, `classifier/`
8. Write fresh `~/.claude/alm/state.json`:
```json
{
"paused": false,
"installDate": "{current ISO timestamp}Z",
"evalCount": 0,
"lastReflection": null,
"firstRunComplete": false
}
```
9. Write empty `~/.claude/alm/confidence.json`: `{}`
### Report
Output: `[ALM] Full reset complete. All learning data has been cleared. Seed playbooks will be used on next session.`From sfw/ALM, MIT.
More from sfw/ALM
- 0.409score
reflect
sfw
Analyze ALM outcomes and corrections to update playbooks and retrain classifier.
SkillDFreshness: DClaude Code +9 - 0.406score
review
sfw
View and correct recent ALM evaluations.
SkillDFreshness: DClaude Code +9 - 0.404score
export
sfw
Export ALM playbooks as standalone markdown for sharing, version control, or manual editing.
SkillDFreshness: DClaude Code +9 - 0.404score
forget
sfw
Hard reset for a specific task type.
SkillDFreshness: DClaude Code +9 - 0.404score
status
sfw
Show ALM learning progress dashboard with session counts, confidence scores, classifier status, and suggested actions.
SkillDFreshness: DClaude Code +9 - 0.399score
resume
sfw
Re-enable ALM evaluation and playbook injection after pausing.
SkillDFreshness: DClaude Code +9 - 0.396score
pause
sfw
Temporarily disable ALM evaluation and playbook injection.
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Feb 7, 2026 (8 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/sfw/ALM
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from sfw
- 0.409score
reflect
sfw
Analyze ALM outcomes and corrections to update playbooks and retrain classifier.
SkillDFreshness: DClaude Code +9 - 0.406score
review
sfw
View and correct recent ALM evaluations.
SkillDFreshness: DClaude Code +9 - 0.404score
export
sfw
Export ALM playbooks as standalone markdown for sharing, version control, or manual editing.
SkillDFreshness: DClaude Code +9 - 0.404score
forget
sfw
Hard reset for a specific task type.
SkillDFreshness: DClaude Code +9 - 0.404score
status
sfw
Show ALM learning progress dashboard with session counts, confidence scores, classifier status, and suggested actions.
SkillDFreshness: DClaude Code +9 - 0.399score
resume
sfw
Re-enable ALM evaluation and playbook injection after pausing.
SkillDFreshness: DClaude Code +9
Cite this page
reset skill. Agent Mart. https://agentmart.dev/skills/sfw-reset (last updated 2026-10-24)