← ClaudeAtlas

ai-audit-loglisted

Audit and log AI interactions into a structured monthly log file. Trigger this skill whenever the user invokes /ai-audit-log or /ai-audit-log --last=N (N: 1–5). The skill reads conversation context to extract prompts and outputs, then asks the user to evaluate each interaction via a structured form. Use this skill to track AI usage quality, model performance, and revision patterns over time.
phatnguyen975/ai-audit-log · ★ 0 · AI & Automation · score 70
Install: claude install-skill phatnguyen975/ai-audit-log
# AI Audit Log Skill Logs AI interactions into default path `docs/audit/ai/<fullname>-YYYY-MM.log.md` with structured entries and monthly statistics. ## Invoke Syntax ```bash /ai-audit-log # log last 1 interaction, save to default path /ai-audit-log --last=N # log last N interactions, save to default path (N: 1–5) /ai-audit-log --file="path/to/output.md" # log last 1 interaction, save to custom file path /ai-audit-log --last=N --file="path/to/output.md" # log last N interactions, save to custom file path (N: 1–5) ``` If `N > 5`, respond: _"Max --last is 5. Running with --last=5."_ and proceed. ## Fullname Resolution The log filename is `<fullname>-YYYY-MM.log.md`. Resolve `fullname` as follows: 1. **Context available** — if human has provided their name earlier in the conversation, use it directly (no need to ask again) 2. **No context** — ask: _"What is your full name? (Used for the log filename)"_ and wait for the answer 3. **Normalization rules** (always applied): - Convert to lowercase - Remove diacritics/accents - Replace spaces with `-` - **Example:** `Nguyễn Tấn Phát` → `nguyen-tan-phat` ## Output Path Resolution | Invocation | Output path | | ---------------------------- | ----------------------------------------- | | No `--file` flag | `docs/audit/ai/<fullname>-YYYY-MM.log.md` | | `--file="path/to/output.md"` | E