{
"schema_version": "2.0",
"meta": {
"generated_at": "2026-07-04T13:57:40.310Z",
"slug": "0xrdan-learn-off",
"source_url": "https://github.com/0xrdan/claude-router/tree/main/.claude-plugin/skills/learn-off",
"source_ref": "main",
"model": "codex",
"analysis_version": "3.0.0",
"source_type": "community",
"content_hash": "5fdd0a211c507c0eda699fe6fa0cb4b8a39570a64e487f434ef6f4086d3dcee1",
"tree_hash": "654a091ea1937a06e2804ed0a66da3ef4750b625eb27d9809719ac0e0eaab267",
"provenance": {
"model_effective": "codex",
"fallback_chain": [
{
"agent": "codex",
"outcome": "success"
}
],
"analysis_version": "3.0.0"
}
},
"skill": {
"name": "learn-off",
"description": "Disable continuous learning mode while keeping existing knowledge and manual learning available.",
"summary": "Turns off automatic insight extraction for the current knowledge workflow.",
"icon": "📦",
"version": "1.0.1",
"author": "0xrdan",
"license": "MIT",
"tags": [
"continuous-learning",
"knowledge-management",
"state-control",
"session-summary"
],
"supported_tools": [
"claude",
"codex",
"claude-code"
],
"risk_factors": [
"external_commands"
],
"category": "productivity"
},
"security_audit": {
"risk_level": "safe",
"is_blocked": false,
"safe_to_publish": true,
"analysis_status": "ok",
"summary": "All seven static findings are false positives caused by Markdown inline code or fenced blocks in SKILL.md. The skill only describes updating a scoped learning-mode state file and does not include executable code, network access, environment access, or prompt-injection text. No additional semantic security findings were found.",
"remediation": [],
"risk_factor_evidence": [
{
"factor": "external_commands",
"evidence": [
{
"file": "SKILL.md",
"line_start": 16,
"line_end": 16
},
{
"file": "SKILL.md",
"line_start": 20,
"line_end": 20
},
{
"file": "SKILL.md",
"line_start": 22,
"line_end": 27
},
{
"file": "SKILL.md",
"line_start": 27,
"line_end": 28
},
{
"file": "SKILL.md",
"line_start": 28,
"line_end": 33
},
{
"file": "SKILL.md",
"line_start": 33,
"line_end": 45
},
{
"file": "SKILL.md",
"line_start": 45,
"line_end": 51
}
]
}
],
"static_findings": [
{
"id": "external_commands:SKILL.md:16:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 16,
"line_end": 16,
"snippet": "- Manual `/learn` commands still work"
},
{
"id": "external_commands:SKILL.md:20:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 20,
"line_end": 20,
"snippet": "1. Read `knowledge/state.json`"
},
{
"id": "external_commands:SKILL.md:22:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 22,
"line_end": 27,
"snippet": "```json"
},
{
"id": "external_commands:SKILL.md:27:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 27,
"line_end": 28,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:28:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 28,
"line_end": 33,
"snippet": "3. Write updated state back to `knowledge/state.json`"
},
{
"id": "external_commands:SKILL.md:33:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 33,
"line_end": 45,
"snippet": "```"
},
{
"id": "external_commands:SKILL.md:45:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 45,
"line_end": 51,
"snippet": "```"
}
],
"finding_verdicts": [
{
"id": "external_commands:SKILL.md:16:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.94,
"reason": "The backticks mark the /learn command in Markdown text. They do not invoke Ruby or shell execution."
},
{
"id": "external_commands:SKILL.md:20:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.94,
"reason": "The backticks identify the knowledge/state.json path in prose. This is a Markdown path reference, not executable shell syntax."
},
{
"id": "external_commands:SKILL.md:22:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "This line opens a fenced JSON example in Markdown. It is documentation formatting and contains no command execution behavior."
},
{
"id": "external_commands:SKILL.md:27:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "This line closes a Markdown code fence. It is not Ruby backtick execution or a shell command."
},
{
"id": "external_commands:SKILL.md:28:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.94,
"reason": "The backticks wrap the knowledge/state.json filename inside an instruction. No shell command or dynamic execution is present."
},
{
"id": "external_commands:SKILL.md:33:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "This line starts a Markdown output-format block. It only documents the expected response text and does not execute commands."
},
{
"id": "external_commands:SKILL.md:45:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "This line ends a Markdown output-format block. The backticks are formatting delimiters, not executable syntax."
}
],
"semantic_findings": [],
"critical_findings": [],
"high_findings": [],
"medium_findings": [],
"low_findings": [],
"dangerous_patterns": [],
"files_scanned": 1,
"total_lines": 52,
"audit_model": "codex",
"audited_at": "2026-07-04T13:57:40.310Z",
"agent_auto_install_policy": "allowed",
"manual_install_policy": "allowed"
},
"content": {
"user_title": "Disable Continuous Learning Mode",
"value_statement": "Continuous learning can keep extracting insights after a focused session ends. This skill turns it off and confirms the learning state clearly.",
"seo_keywords": [
"Claude",
"Codex",
"Claude Code",
"learn-off skill",
"continuous learning mode",
"knowledge state",
"AI skill marketplace",
"manual learn command",
"learning mode toggle"
],
"actual_capabilities": [
"Sets learning_mode to false in knowledge/state.json.",
"Clears learning_mode_since by setting it to null.",
"Stops automatic insight extraction for future interactions.",
"Stops query counting while continuous learning is disabled.",
"Confirms the disabled state with a short session summary.",
"Keeps manual /learn extraction available."
],
"limitations": [
"Does not delete existing captured insights.",
"Does not disable manual /learn commands.",
"Requires knowledge/state.json to be available for state updates.",
"Does not re-enable learning mode; use /learn-on for that action."
],
"use_cases": [
{
"title": "End a focused learning session",
"description": "Turn off automatic insight extraction when a research or coding session is complete.",
"target_user": "Individual developers"
},
{
"title": "Pause workspace insight capture",
"description": "Stop background learning before switching to unrelated tasks or sensitive discussion.",
"target_user": "Team leads"
},
{
"title": "Keep manual knowledge capture",
"description": "Disable automatic capture while preserving manual commands for selected insights.",
"target_user": "Documentation maintainers"
}
],
"prompt_templates": [
{
"title": "Turn Learning Off",
"scenario": "Beginner request for the default action.",
"prompt": "Disable continuous learning mode now and show the status summary."
},
{
"title": "Preserve Existing Knowledge",
"scenario": "Confirm that stored insights remain available.",
"prompt": "Disable continuous learning, preserve existing knowledge, and tell me whether manual /learn remains available."
},
{
"title": "Review Session Counts",
"scenario": "Request a state change with available metrics.",
"prompt": "Audit the current learning state, switch continuous learning off, and provide extraction, query, and insight counts if present."
},
{
"title": "Close A Project Review",
"scenario": "Advanced request for a precise state update.",
"prompt": "Disable learning mode after this project review, update only the learning state, and give a concise completion note with any missing count fields called out."
}
],
"output_examples": [
{
"input": "Disable continuous learning mode.",
"output": "Continuous Learning: DISABLED. Learning mode is now inactive. Manual extraction remains available through /learn."
},
{
"input": "Turn learning off and summarize the session counts.",
"output": "Continuous learning is disabled. The summary lists available extraction, query, and insight counts, and notes any values that were not recorded."
},
{
"input": "Stop automatic learning but keep existing knowledge.",
"output": "Automatic insight extraction is off. Existing knowledge remains available, and the knowledge base can still be viewed separately."
}
],
"best_practices": [
"Use this skill when a focused learning session is complete.",
"Review the summary to confirm the learning state changed as expected.",
"Use manual /learn for selected insights after automatic capture is disabled."
],
"anti_patterns": [
"Do not use this skill when you want continuous learning to keep running.",
"Do not expect this skill to remove stored knowledge.",
"Do not use it as a replacement for reviewing sensitive workspace data policies."
],
"faq": [
{
"question": "Does this delete existing knowledge?",
"answer": "No. It only disables continuous learning mode and keeps captured insights available."
},
{
"question": "Can I still use /learn after running it?",
"answer": "Yes. Manual /learn commands remain available for selected insight capture."
},
{
"question": "What file does it update?",
"answer": "The skill instructs the assistant to update knowledge/state.json with the disabled learning state."
},
{
"question": "Does it stop query counting?",
"answer": "Yes. The skill states that query counting stops when continuous learning is disabled."
},
{
"question": "How do I turn learning back on?",
"answer": "Use the related /learn-on command when you want automatic learning to resume."
},
{
"question": "Is this skill for Claude, Codex, and Claude Code?",
"answer": "Yes. The report lists support for Claude, Codex, and Claude Code."
}
]
},
"file_structure": [
{
"name": "SKILL.md",
"type": "file",
"path": "SKILL.md",
"lines": 52
}
]
}