메뉴
Reads text content from PDF files. Use when you need to extract text, analyze content, or summarize information from a PDF document. Supports limiting the number of pages read.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Create or edit SVG diagrams that must remain editable in diagrams.net/draw.io or the drawio-obsidian plugin, instead of becoming plain static SVG images.
Explain a codebase in four modes: `framework` produces a high-level architecture walkthrough saved to `codebase-walkthrough[-<slug>].md`; `impl <path>` explains a specific implementation saved to `codebase-impl-<slug>.md`; `run <identifier>` explains a specific execution run by tracing code + logs, saved to `codebase-run-<slug>.md`; `debug [<identifier>] [<question>]` diagnoses failures and verifies fixes, saved to `codebase-debug-<slug>.md`. `<slug>` is a short (≤12 chars) tag derived from the identifier so repeated runs don't overwrite each other.
Analyzes the current AXRD tmux session by using the shared TMA mechanism from $HOME/farside/AGENTS.md: one top-level summary TMA Agent summarizes the run, and one workspace TMA analysis Agent deeply analyzes each AXRD workspace.
Create or refine concise skills for task-specific TMA Agent collaboration systems. Use when the user wants to turn a tmux multi-agent workflow into a reusable skill, or update an existing TMA skill based on what happened during one of its invocations. Shared TMA creation/communication conventions live in $HOME/farside/AGENTS.md.
Respawns tmux AI agent panes so they reload current authentication while preserving Codex sessions when possible. Use when Codex panes are logged into an old account, stale after auth changes, or the user asks to refresh all AI windows in tmux.
For repo update tasks, pull the latest code first, then continue with the repo-appropriate local configuration or environment updates.
| name | pdf-reader |
| description | Reads text content from PDF files. Use when you need to extract text, analyze content, or summarize information from a PDF document. Supports limiting the number of pages read. |
This skill allows you to extract and read text content from PDF files using pypdf.
To read a PDF file, execute the following command using uv:
XDG_CACHE_HOME=/tmp/.cache UV_CACHE_DIR=/tmp/.uv-cache \
uv run --with pypdf python3 ~/deploy/configs/llm/skills/pdf-reader/scripts/read_pdf.py <path_to_pdf> [--pages <number_of_pages>]
Read an entire PDF:
XDG_CACHE_HOME=/tmp/.cache UV_CACHE_DIR=/tmp/.uv-cache \
uv run --with pypdf python3 ~/deploy/configs/llm/skills/pdf-reader/scripts/read_pdf.py /path/to/document.pdf
Read only the first 5 pages:
XDG_CACHE_HOME=/tmp/.cache UV_CACHE_DIR=/tmp/.uv-cache \
uv run --with pypdf python3 ~/deploy/configs/llm/skills/pdf-reader/scripts/read_pdf.py /path/to/document.pdf --pages 5
If uv fails because it cannot write to ~/.cache/uv (common in sandboxed sessions), re-run with temporary cache directories:
XDG_CACHE_HOME=/tmp/.cache UV_CACHE_DIR=/tmp/.uv-cache \
uv run --with pypdf python3 ~/deploy/configs/llm/skills/pdf-reader/scripts/read_pdf.py <path_to_pdf> [--pages <number_of_pages>]
If uv is not available, ensure pypdf is installed via pip:
pip install pypdf
python3 ~/deploy/configs/llm/skills/pdf-reader/scripts/read_pdf.py <path_to_pdf> --pages <number_of_pages>