/deep-execution

name: deep-execution
description: Executes agent-enhanced council queries as one Workflow of parallel Claude analyst agents that each query a provider, evaluate response quality, ask follow-up questions, and return a schema-enforced analysis with confidence ratings and blind spot analysis. Invoked when the --agents flag is used or when complex architectural decisions are detected.

Agent-Enhanced Council Execution

Run one Workflow of parallel Claude analyst agents for deeper analysis. Each analyst queries its provider, evaluates response quality, can ask follow-up questions, and returns a structured analysis that the workflow enforces against the schema.

Step 1: Resolve Providers and Write the Questions

One shell call resolves what the workflow needs — each provider's model, its role, and the question file it will send — with the same helpers the standard flow uses. Paste the final question into the heredoc verbatim: the quoted marker means the shell does NOT interpret quotes, backticks or `$()` in it. The question is emitted once; each provider's role-injected variant is built in shell, not pasted again.

source "${CLAUDE_PLUGIN_ROOT}/scripts/lib/providers.sh"
source "${CLAUDE_PLUGIN_ROOT}/scripts/lib/roles.sh"
PROVIDERS=(<the selected providers, space separated>)
ROLES="<the --roles value, or empty>"
RUN=$(date +%s)                                  # names this run's files; Step 6 reuses it
Q="$PWD/.claude/council-cache/.agents-$RUN"
mkdir -p "$PWD/.claude/council-cache"
# Self-ignoring, as cache.sh and run-council.sh keep it: these files carry the
# question and any --file contents, and must never land in a commit.
[[ -f "$PWD/.claude/council-cache/.gitignore" ]] || printf '*\n' > "$PWD/.claude/council-cache/.gitignore"
cat > "$Q.txt" <<'COUNCIL_Q_EOF'
<the question, verbatim>
COUNCIL_Q_EOF
# --file / auto-context: append the context to the question file here, e.g.
#   { printf '\n\nHere is the content of %s:\n\n```\n' "<path>"; cat "<path>"; printf '```\n'; } >> "$Q.txt"
# An unknown role is refused, as the standard flow refuses it; assigning it
# would hand that provider the bare question under a role heading.
[[ -z "$ROLES" ]] || validate_roles "$ROLES" || exit 1
# The model the ANALYST agents run on — distinct from the provider models below,
# which are what actually answer the question. Pinned rather than inherited: an
# omitted model makes every analyst run on whatever model the user's session
# happens to use, so the cost of a mode that already fans out one agent per
# provider varies by a factor of several for no stated reason. Sonnet handles
# the analyst's work — run a query, judge the answer, emit a structured object.
ANALYST_MODEL="${COUNCIL_AGENT_MODEL:-sonnet}"
case "$ANALYST_MODEL" in
    sonnet|opus|haiku|fable) ;;
    *) echo "COUNCIL_AGENT_MODEL must be one of: sonnet, opus, haiku, fable (got '$ANALYST_MODEL')" >&2; exit 1 ;;
esac
echo "analyst model $ANALYST_MODEL"
ASSIGNMENTS=""
[[ -n "$ROLES" ]] && ASSIGNMENTS=$(assign_roles_to_providers "$ROLES" "${PROVIDERS[@]}")
echo "run $RUN"
for p in "${PROVIDERS[@]}"; do
    role=""
    [[ -n "$ASSIGNMENTS" ]] && role=$(get_provider_role "$p" "$ASSIGNMENTS")
    qf="$Q.txt"
    if [[ -n "$role" ]]; then
        qf="$Q-$p.txt"
        build_prompt_with_role "$(cat "$Q.txt")" "$role" > "$qf"
    fi
    printf '%s\t%s\t%s\n' "$p" "$(get_model "$p")" "$qf"
done

It prints the run id, then the analyst model, then one line per provider: name, model (shown in the Step 4 header), and the absolute path of that provider's question file. Pass the analyst model to the workflow as `analystModel`; the workflow script cannot read the environment itself.

Step 2: Run the Analyst Workflow

Agent mode is one Workflow: one analyst agent per provider, in parallel, each returning its analysis through schema-enforced structured output. The user asked for agent mode (`--agents`, or yes to the prompt in ask.md Step 1.5), which is the opt-in the Workflow tool requires.

If the Workflow tool is not available in this session, stop here: tell the user that `--agents` needs a Claude Code with the Workflow tool, and offer to run the same question in standard mode instead. Do not fall back to another way of spawning agents.

Read `${CLAUDE_PLUGIN_ROOT}/schemas/agent-analysis.schema.json` with the Read tool, then call the Workflow tool with this script, verbatim, via `script`:

export const meta = {
  name: 'council-agents',
  description: 'One analyst per council provider: query it, judge the answer, follow up, return a structured analysis',
  phases: [{ title: 'Analyze', detail: 'one analyst agent per provider, in parallel' }],
}
// The tool's schema validator does not know the draft-2020-12 dialect the
// file declares; without the declaration it validates the same keywords fine.
const schema = { ...args.schema }
delete schema.$schema
const template = `${args.pluginRoot}/skills/deep-execution/agent-prompt-template.md`
// Single stage: there is no later step for a finished analyst to move on to,
// so the barrier costs nothing.
const results = await parallel(args.providers.map(p => () =>
  agent(
    `You are the council analyst for the provider "${p.name}".\n` +
    `Read ${template} and carry out every step in it, with these values:\n` +
    `- {PROVIDER} = ${p.name}\n` +
    `- {PLUGIN_ROOT} = ${args.pluginRoot}\n` +
    `- {QUESTION_FILE} = ${p.questionFile}\n` +
    `Your final answer is the Round 3 analysis object, returned through the structured output tool.`,
    { label: p.name, phase: 'Analyze', schema, agentType: 'general-purpose',
      model: args.analystModel })))
// parallel() keeps a dead or skipped analyst's slot as null, index-aligned with args.providers.
const failed = args.providers.filter((_, i) => !results[i]).map(p => p.name)
if (failed.length) log(`no analysis from: ${failed.join(', ')}`)
return {
  // The label goes last so an analyst that emits its own "provider" key cannot rename itself.