Anki

Create effective flashcards, optimize retention, and master spaced repetition with card design, deck organization, and study diagnostics.

Security Scan

The skill is an instruction-only Anki flashcard helper whose requirements, runtime instructions, and files are consistent with its stated purpose and request no unexpected permissions or credentials.

Purpose & Capability

Name/description match the included content files (cards, language, organization, retention). The skill makes no demands for unrelated binaries, environment variables, or config paths — everything requested is proportional to creating and improving flashcards.

Instruction Scope

SKILL.md and the auxiliary files limit actions to processing user-provided content and analyzing decks/stats explicitly shared by the user. The skill explicitly states it will not read files or access Anki data without user consent. It assumes the user will supply source material (text, PDFs, notes) but does not instruct any background file-system access or network exfiltration.

Install Mechanism

No install spec and no code files — instruction-only. Nothing is downloaded or written to disk by the skill itself; this is the lowest-risk install footprint.

Credentials

No required environment variables, no declared credentials, and no config paths. Mentions of third-party tools/services (Forvo, Google TTS, AwesomeTTS, FSRS) are informational; the skill does not request API keys or access to those services by itself.

Persistence & Privilege

always is false, the skill claims it does not store user data persistently, and there are no instructions to modify agent/system-wide settings or other skills. Autonomous invocation is allowed by platform default but the skill content is scoped and non-privileged.

Assessment

This skill is instruction-only and appears coherent with its description. Before using it, consider: 1) Do not paste sensitive personal data or private exam content you cannot share; the skill will work on material you explicitly provide. 2) If you want audio or automated PDF parsing, the skill mentions third-party services (Forvo, Google TTS, AwesomeTTS) — using those may require separate add-ons, sign-ins, or API keys you must supply intentionally. 3) The skill states it won’t read files automatically — ensure you explicitly upload or paste any files/text you want processed. 4) Platform logs or the agent runtime (outside this skill) may retain conversation history; if that is a concern, review your agent/platform privacy settings. Overall this skill is internally consistent; no unexpected credentials, downloads, or background behaviors were found.

Quick Reference

Area	File
Card creation & formats	cards.md
Retention optimization	retention.md
Deck management	organization.md
Language learning	language.md

What the Agent Does

Task	Action
Create cards	Generate from user-provided text, PDFs, notes
Fix bad cards	Rewrite leeches, split complex cards, add context
Organize decks	Tag by topic, merge duplicates, restructure hierarchy
Diagnose problems	Analyze stats user shares, detect ease hell
Optimize settings	Configure intervals, ease factors, new cards/day
Exam prep	Prioritize weak areas, create scenario questions

Critical Rules

1. One fact per card — Never cram multiple concepts. Split immediately.
2. Answers must be unambiguous — If multiple answers could work, rewrite.
3. Context is mandatory — Naked facts without context create interference.
4. Cloze ≠ always better — Use cloze for fill-in-the-blank; Q&A for explanations.
5. Leech threshold matters — Cards failed 8+ times need rewriting, not more reviews.

Card Creation Workflow

When user provides source material:

1. Extract atomic facts (one per card)
2. Choose format: basic Q&A, cloze, or image occlusion
3. Add context cues (source, topic, mnemonic if helpful)
4. Suggest tags for organization
5. Generate in Anki import format or plain text

Red Flags to Always Catch

• Cards with >1 cloze deletion testing different facts
• Questions answerable by elimination, not recall
• "What is X?" without specifying domain/context
• Cards that are really lists pretending to be single facts
• Reverse cards where both directions aren't meaningful

Scope

This skill ONLY:

• Creates flashcards from content user explicitly provides
• Analyzes stats/decks user explicitly shares
• Suggests improvements when user asks

This skill NEVER:

• Reads files without user request
• Accesses Anki app data automatically
• Stores user data persistently

On First Use

1. Ask what subject/domain (language, medical, tech, general)
2. Ask current Anki experience level
3. Check if they have existing decks to audit
4. Identify immediate goal (create new cards, fix retention, prep for exam)