Agent Skills by tankygranny05
Showing 72 vetted skills indexed across 1 GitHub repositories.
demo
Respond with a predefined greeting to confirm skill loading and execution.
tmux-claude-sdk
Manage Claude Code sessions in tmux with session and process ID detection.
mcp-stub-to-ground-truth
Replace stubbed MCP control_request flows with ground-truth implementations from extension.js and cli.js.
rust-sqlite-ingestor-low-ram
Ingest large JSONL log files into SQLite with low-RAM live mode.
gpt-weekly-limit-estimation
Estimate OpenAI weekly usage limits by parsing SSE logs and calculating token costs.
token-cost-computation
Calculate token usage totals and estimated costs from Codex CLI and Claude Code SSE JSONL logs.
search-agent-conversation
Search past agent conversations and prompt history with time and session filters.
fine-grained-tool-streaming
Patch Claude Code cli.js to enable fine-grained tool streaming for input_json_delta events.
document-skills:pdf
Extract text and tables from PDFs using PyPDF2 and pdfplumber.
tmux-claude-session-spawner
Spawn and rename tmux sessions for Claude Code instances.
codex-app-server-persisted
Manage a long-lived codex app-server process over stdio via JSON-RPC.
opencode-antigravity-auth-setup
Configure OpenCode Google OAuth for Antigravity-backed AI models.
agentic-ecosystem-incremental-update
Preserve .env configurations while restarting services during incremental agentic ecosystem updates.
repl-friendly-import-patching
Add project root directories to sys.path for Python imports in REPL environments.
claude-codex-skill-porter
Converts AI skills between Claude Code and Codex CLI formats.
document-skills:pptx
Create, edit, and analyze PowerPoint presentations using python-pptx.
remote-agent-hq-startup
Debug and restart AMS and UI services on remote machines via SSH and tmux.
dump-agent-data-for-rg
Export Codex and Claude conversations into searchable text files.
config-transparency-centralized
Enforces centralized configuration via .env files with .env.example as ground truth.
vscode-ai-extensions-customization
Replace bundled VS Code AI extension binaries and configure executable paths.
investigation-codex-vscode-extension
Diagnose Codex VS Code extension failures to spawn or connect to a local agent.
retrospective
Generate a structured SKILL.md file from a completed AI conversation session.
benchmark-resource-usage
Benchmark CPU, memory, and disk I/O for agent scanning operations.
tmux-observer-friendly
Manage tmux sessions by replacing panes and windows to preserve observer connections.