Skills for Real Engineers. Straight from my .agents directory.
council-execution
Agent skill fromhex/claude-council
Executes council queries by running the query pipeline across selected AI providers (Gemini, OpenAI, Grok, Perplexity), displaying formatted responses verbatim, and generating a synthesis of consensus, divergence, and recommendations. Invoked by the ask command during standard (non-agent) council queries.
How to install council-execution
npx skills add hex/claude-council# or add only this skill to Claude Code: copy the folder
skills/council-execution/
# into
~/.claude/skills/council-execution/ (all projects)
.claude/skills/council-execution/ (one project)
1# Council Query Execution23## Step 1: Run Query and Save to File45```bash6bash ${CLAUDE_PLUGIN_ROOT}/scripts/run-council.sh --providers=gemini,openai -- "Your question"7```89This outputs the path to the saved file (e.g., `.claude/council-cache/council-1734567890.md`).1011**Flag syntax**: Use `=` with no spaces: `--providers=gemini,openai`1213**Pass only the flags the user asked for.** Never add `--no-pane`, `--quiet`,14`--no-cache` or `--no-auto-context` on your own judgment; they are the user's15to choose, and the script already skips a pane that cannot open.1617**CRITICAL**: Always place `--` before the prompt to prevent prompt text containing dashes from being parsed as flags.1819**Run this Bash call with `timeout: 600000`** (ten minutes). The Bash tool's20two-minute default is shorter than a slow council round, and inside tmux a21provider failure makes the streaming pane offer the user a retry for22`COUNCIL_RETRY_WAIT` seconds (default 45) and then runs another provider round23if accepted. A call cut off mid-run loses the whole transcript, the answers24that did arrive included.2526## Step 2: Read and Display the Output VERBATIM2728Use the **Read tool** to read the output file path returned by Step 1.2930**If the file's first line is `<!-- council: round 1 was shown in the pane -->`**,31the user already read every round-1 answer in the streaming pane. Do not32reprint them. Instead, output one line per provider in round 1, taken from33…
Excerpt shown under the repository's MIT license.
More skills in claude-council
Similar skill repositories
An agentic skills framework & software development methodology that works.
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
Public repository for Agent Skills
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Production-grade engineering skills for AI coding agents.
