terrylica/cc-skills

summarize

AI-powered iterative deep-dive analysis of converted recordings.

ソースを見る

使い方

  1. インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
  2. 必要なツール、認証情報、対応環境を原文で確認します。
  3. 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/terrylica/cc-skills --skill summarize

開始前の確認

  • 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
  • スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
リポジトリの原文

見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。

/asciinema-tools:summarize

AI-powered iterative deep-dive analysis for large .txt recordings. Uses guided sampling and AskUserQuestion to progressively explore the content.

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

Philosophy

Large recordings (1GB+) cannot be read entirely. This command uses:

  1. Initial guidance - What are you looking for?
  2. Strategic sampling - Head, middle, tail + keyword-targeted sections
  3. Iterative refinement - AskUserQuestion to drill deeper into findings
  4. Progressive synthesis - Build understanding through multiple passes

Arguments

ArgumentDescription
filePath to .txt file (converted from .cast)
--topicInitial focus area (e.g., "ML training", "errors")
--depthAnalysis depth: quick, medium, deep
--outputSave findings to markdown file

Workflow

Phase 1: Initial Guidance

yaml
AskUserQuestion:
  question: "What are you trying to understand from this recording?"
  header: "Focus"
  options:
    - label: "General overview"
      description: "What happened in this session? Key activities and outcomes"
    - label: "Key findings/decisions"
      description: "Important discoveries, conclusions, or decisions made"
    - label: "Errors and debugging"
      description: "What went wrong? How was it resolved?"
    - label: "Specific topic"
      description: "I'll specify what I'm looking for"

Phase 2: File Statistics

bash
/usr/bin/env bash << 'STATS_EOF'
FILE="$1"

echo "=== File Statistics ==="
SIZE=$(ls -lh "$FILE" | awk '{print $5}')
LINES=$(wc -l < "$FILE")
echo "Size: $SIZE"
echo "Lines: $LINES"

echo ""
echo "=== Content Sampling ==="
echo "First 20 lines:"
head -20 "$FILE"

echo ""
echo "Last 20 lines:"
tail -20 "$FILE"

echo ""
echo "=== Keyword Density ==="
echo "Errors/failures:"
grep -c -i "error\|fail\|exception" "$FILE" || echo "0"
echo "Success indicators:"
grep -c -i "success\|complete\|done\|pass" "$FILE" || echo "0"
echo "Key decisions:"
grep -c -i "decision\|chose\|selected\|using" "$FILE" || echo "0"
STATS_EOF

Phase 3: Strategic Sampling

Based on file size, sample strategically:

For files < 100MB:

bash
# Sample head, middle, tail (1000 lines each)
head -1000 "$FILE" > /tmp/sample_head.txt
tail -1000 "$FILE" > /tmp/sample_tail.txt
TOTAL=$(wc -l < "$FILE")
MIDDLE=$((TOTAL / 2))
sed -n "${MIDDLE},$((MIDDLE + 1000))p" "$FILE" > /tmp/sample_middle.txt

For files > 100MB:

bash
# Keyword-targeted sampling
grep -B5 -A20 -i "$TOPIC_KEYWORDS" "$FILE" | head -5000 > /tmp/sample_targeted.txt

Phase 4: Initial Analysis

Read the samples and provide initial findings. Then ask:

yaml
AskUserQuestion:
  question: "Based on initial analysis, what would you like to explore deeper?"
  header: "Drill down"
  multiSelect: true
  options:
    - label: "Specific timeframe"
      description: "Jump to a particular section (e.g., 'around line 50000')"
    - label: "Follow keyword trail"
      description: "Search for specific patterns and expand context"
    - label: "Error investigation"
      description: "Deep dive into errors and their resolution"
    - label: "Success moments"
      description: "What worked? What were the wins?"
    - label: "Generate summary"
      description: "Synthesize findings into a report"

Phase 5: Iterative Deep-Dive

For each selected focus area:

  1. Extract relevant sections using grep with context
  2. Read and analyze the extracted content
  3. Report findings to user
  4. Ask for next action via AskUserQuestion
yaml
AskUserQuestion:
  question: "Found {N} relevant sections. What next?"
  header: "Continue"
  options:
    - label: "Show me the most significant"
      description: "Display top 3 most relevant excerpts"
    - label: "Search for related patterns"
      description: "Expand search to related keywords"
    - label: "Move on"
      description: "I have enough on this topic"

Phase 6: Synthesis

yaml
AskUserQuestion:
  question: "Ready to generate summary. What format?"
  header: "Output"
  options:
    - label: "Concise bullet points"
      description: "Key findings in 10-15 bullets"
    - label: "Detailed markdown report"
      description: "Full report with sections and evidence"
    - label: "Executive summary"
      description: "1-paragraph high-level summary"
    - label: "Save to file"
      description: "Write findings to markdown file"

Keyword Libraries

Trading/ML Domain

sharpe|drawdown|backtest|overfitting|regime|validation
model|training|loss|epoch|gradient|convergence
feature|indicator|signal|position|portfolio

Development Domain

error|exception|fail|bug|fix|debug
commit|push|merge|branch|deploy
test|assert|verify|validate|check

Claude Code Domain

tool|bash|read|write|edit|grep
task|agent|subagent|spawn
permission|approve|reject|block

Example Usage

bash
# Interactive exploration
/asciinema-tools:summarize session.txt

# Focused on ML findings
/asciinema-tools:summarize session.txt --topic "ML training results"

# Quick overview
/asciinema-tools:summarize session.txt --depth quick

# Full analysis with report
/asciinema-tools:summarize session.txt --depth deep --output findings.md

Example Output

markdown
# Session Summary: research-repo-research_20251226

## Overview

- **Duration**: 4 days (Dec 26-30, 2025)
- **Size**: 12GB recording → 3.2GB text
- **Primary Focus**: ML robustness research

## Key Findings

### 1. Training-Evaluation Mismatch (CRITICAL)

- MSE loss optimizes magnitude, but Sharpe evaluates direction
- Result: 80% Sharpe collapse from 2024 to 2025

### 2. Fishr λ=0.1 Solution (BREAKTHROUGH)

- Gradient variance penalty solves V-REx binary threshold
- Feb'24 Sharpe: -6.14 → +6.14

### 3. Model Rankings

| Model  | Window | Sharpe |
| ------ | ------ | ------ |
| TFT    | 15mo   | 1.02   |
| BiLSTM | 12mo   | 0.50   |

## Evidence Locations

- Line 15234: "Fishr λ=0.1 SOLVES the V-REx binary threshold problem"
- Line 48102: Phase 4 results summary table

## Next Steps Identified

1. TFT 15mo + Fishr training
2. PSR/PBO statistical validation
3. Agent research synthesis

Troubleshooting

IssueCauseSolution
File too largeRecording exceeds memory limitUse --depth quick for sampling only
No keywords foundWrong domain or sparse contentTry different --topic focus area
Sampling timeoutVery large fileIncrease terminal timeout or use grep
grep context errorMissing GNU grepbrew install grep (BSD grep limits)
Output file not savedPermission deniedCheck write permissions on --output

Related Commands

  • /asciinema-tools:convert - Convert .cast to .txt first
  • /asciinema-tools:analyze - Keyword-based analysis (faster, less deep)
  • /asciinema-tools:finalize - Process orphaned recordings

Post-Execution Reflection

After this skill completes, reflect before closing the task:

  1. Locate yourself. — Find this SKILL.md's canonical path before editing.
  2. What failed? — Fix the instruction that caused it.
  3. What worked better than expected? — Promote to recommended practice.
  4. What drifted? — Fix any script, reference, or dependency that no longer matches reality.
  5. Log it. — Evolution-log entry with trigger, fix, and evidence.

Do NOT defer. The next invocation inherits whatever you leave behind.

同じリポジトリから

関連する Skills

すべての Skills
terrylica
コミュニティ

academic-pdf-to-gfm

Convert academic PDF papers to GitHub-renderable GFM markdown with math equations. TRIGGERS - PDF, GitHub markdown, math

1時間の導入数
1
GitHub Stars
2
更新日
11月5日
terrylica
コミュニティ

adr-code-traceability

Add ADR references to code for traceability. TRIGGERS - ADR traceability, code reference, document decision in code.

1時間の導入数
1
GitHub Stars
2
更新日
11月5日
terrylica
コミュニティ

arxiv-source-first-paper-ingest

Read an arXiv paper faithfully by fetching the authors' own LaTeX source instead of OCR-ing the PDF. Extracts every display equation with its label, environment and enclosing proposition, renders each to MathML, and proves usability by compiling each formula with a real TeX engine. Use when you need a paper's formulas exactly — to replicate a method, implement a loss function or custom objective, verify a quoted number, or check a claim against what the authors actually wrote. TRIGGERS - read th

1時間の導入数
1
GitHub Stars
2
更新日
11月5日
terrylica
コミュニティ

azure-provision

Bootstrap Azure automation when interactive az login is blocked by a Conditional Access policy (error 530035, "Device state: Unregistered"). Mints a service principal through the Azure Portal session ONCE — using Microsoft Graph over a captured token, not DOM clicking — then provisions resource groups, Cognitive Services / AI Language / Speech resources and their keys entirely via az/REST, writing key AND endpoint together into the SCS vault. Use when az login fails but the portal works, when mo

1時間の導入数
1
GitHub Stars
2
更新日
11月5日