session:extract

Documents

session:extract is an Agent Skill for documents published in the kagenti/kagenti repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Extract Claude Code session analytics to CSV, Markdown, or HTML files

View on GitHub ↗

QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

Open your project in Codex.
Copy and paste the prompt below.
Review files and risks before approving any action.

Prompt to paste

Inspect the Agent Skill at https://github.com/kagenti/kagenti/blob/HEAD/.claude/skills/session:extract/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/session-extract, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

flowchart TD
    START(["/session:extract"]) --> DETECT["Detect repo"]
    DETECT --> PARAMS{"Date range specified?"}
    PARAMS -->|Yes| EXTRACT["--phase extract with date range"]:::phase
    PARAMS -->|No| EXTRACTALL["--phase extract (all data)"]:::phase
    EXTRACT --> OUTPUT["Report output file locations"]:::done
    EXTRACTALL --> OUTPUT
    classDef phase fill:#2196F3,stroke:#333,color:white
    classDef done fill:#4CAF50,stroke:#333,color:white

Follow this diagram as the workflow.

Extract Session Analytics

Extract Claude Code session analytics from GitHub PR/issue comments into CSV, Markdown, or HTML files for offline analysis and reporting.

Workflow

Step 1: Detect Repository

git remote get-url origin | sed 's/.*github.com[:/]\(.*\)\.git/\1/' | sed 's/.*github.com[:/]\(.*\)/\1/'

Step 2: Determine Date Range

Ask the user or accept parameters:

--from-date - Start date (ISO format: YYYY-MM-DD), defaults to 30 days ago
--to-date - End date (ISO format: YYYY-MM-DD), defaults to today

Step 3: Run the Extract Phase

python3 .claude/scripts/session-analytics.py \
  --phase extract \
  --repo <OWNER/NAME> \
  --from-date <START_DATE> \
  --to-date <END_DATE> \
  --output-dir /tmp/kagenti/session/

This will:

Fetch session comments from all PRs/issues in the date range
Parse session metrics from each comment
Generate output files in /tmp/kagenti/session/:
- sessions.csv - Raw session data in CSV format
- sessions.md - Formatted Markdown report
- dashboard.html - Interactive HTML dashboard

Step 4: Report Output

After extraction completes, report the output file locations to the user:

ls -la /tmp/kagenti/session/

Provide the full paths so the user can open or share the files.

Parameters

Parameter	Source	Required	Default
`--phase`	Always `extract`	Yes	-
`--repo`	Auto-detected from git remote	Yes	-
`--from-date`	User-specified or default	No	30 days ago
`--to-date`	User-specified or default	No	Today
`--output-dir`	Fixed	No	`/tmp/kagenti/session/`

Examples

# Extract last 30 days (default)
python3 .claude/scripts/session-analytics.py \
  --phase extract --repo kagenti/kagenti \
  --output-dir /tmp/kagenti/session/

# Extract specific date range
python3 .claude/scripts/session-analytics.py \
  --phase extract --repo kagenti/kagenti \
  --from-date 2025-01-01 --to-date 2025-01-31 \
  --output-dir /tmp/kagenti/session/

# Extract all time
python3 .claude/scripts/session-analytics.py \
  --phase extract --repo kagenti/kagenti \
  --from-date 2024-01-01 \
  --output-dir /tmp/kagenti/session/

Output Files

File	Format	Use Case
`sessions.csv`	CSV	Import into spreadsheets, data analysis tools
`sessions.md`	Markdown	Share in docs, wikis, or other PRs
`dashboard.html`	HTML	Visual overview, open in browser

Related Skills

session - Router skill for all session analytics
session:post - Post session stats to PR/issue comment
session:summary - Update pinned summary comment
session:dashboard - Generate HTML dashboard