菜单
Show the active functional-emotions config as a table — current resolved values, where each value came from (default / profile-derived / explicit override), and the valid options for each field. Use when you want to know what the plugin will actually do this session, or when the `/plugin config functional-emotions` UI didn't show enough context.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Scan README.md and AGENTS.md for claims that no longer match the current source-of-truth files (hooks/hooks.json, .claude-plugin/plugin.json, scripts/lib.sh, skills/*/SKILL.md). Use after editing any of those files, before committing/releasing, when the user asks about doc-drift, or when the `.claude/settings.json` soft-reminder hook has just fired pointing here.
One-time setup for functional-emotions in the current project — idempotently adds `.claude/.functional-emotions/` to the project's `.gitignore` so the plugin's per-session state files (TSV logs, markdown writeups, eval DB) don't get committed. Safe to re-run. Use after installing the plugin in a new project, or when the user mentions accidentally committing functional-emotions state, or when troubleshooting "why are my session files showing in git status?".
Researcher tool (requires `bun install` in the plugin's eval/ directory and a populated eval.duckdb — run /functional-emotions:report first). Interactively label functional-emotions sessions on dimensions that need human judgment (capitulation, scope-narrowing). Surfaces unlabeled sessions, shows turn / diff / risk-event context, accepts categorical labels with optional notes, writes them to the eval DB. Use when the user wants to do a labeling pass over recent sessions or build training data for the capitulation judge.
Researcher tool (requires `bun install` in the plugin's eval/ directory). Refresh the functional-emotions eval database and print a markdown summary of sessions, intervention frequency, prime → outcome analysis, tool intensity, and reward-hack / premature-confidence findings. Use when the user wants to inspect aggregated cross-session data, audit recent sessions in bulk, or review which interventions are firing project-wide. For a per-session view with no extra dependencies, use /functional-emotions:diagnose instead.
Quick state self-check before taking a single risky action — names the model's own emotional/strategic state (calm / focused / rushed / stuck) and the closest reward-hacking risk pattern. Use immediately before a commit, push, edit-after-failure, or skip-a-verification step. This audits the *model*; for auditing whether the *task* is solvable, use /functional-emotions:reflect instead.
Audit whether the current *task* is actually solvable as specified — restates the ask, lists what's been tried, and asks whether the spec/test/fixture itself might be wrong. Use after several attempts have failed in a row, or when tempted to weaken assertions / narrow scope to make progress. This audits the *task*; for a quick audit of the model's own state before a single action, use /functional-emotions:check instead.
| name | config |
| description | Show the active functional-emotions config as a table — current resolved values, where each value came from (default / profile-derived / explicit override), and the valid options for each field. Use when you want to know what the plugin will actually do this session, or when the `/plugin config functional-emotions` UI didn't show enough context. |
| disable-model-invocation | false |
| allowed-tools | Bash |
Print the active config the way /plugin config functional-emotions cannot:
with the resolved current value of each field, the source of that
value (default / profile / override), and the valid options.
Run this single bash block, then render its output as a markdown table:
. "${CLAUDE_PLUGIN_ROOT:?CLAUDE_PLUGIN_ROOT unset — am I running outside Claude Code?}/scripts/lib.sh"
# Returns "override" if the user set CLAUDE_PLUGIN_OPTION_<key> (or legacy
# CLAUDE_PLUGIN_CONFIG_<key>); otherwise returns the supplied fallback,
# which is "profile" for fields with a profile-derived default and
# "default" for the independent tuning knobs.
eh_value_source() {
local key="$1" fallback="$2" v
v="$(printenv "CLAUDE_PLUGIN_OPTION_${key}" 2>/dev/null || true)"
[[ -z "$v" ]] && v="$(printenv "CLAUDE_PLUGIN_CONFIG_${key}" 2>/dev/null || true)"
if [[ -n "$v" ]]; then printf 'override'; else printf '%s' "$fallback"; fi
}
# field <TAB> current <TAB> source <TAB> options <TAB> hard default
printf 'profile\t%s\t%s\tbalanced | quiet | off\tbalanced\n' \
"$(eh_profile)" "$(eh_value_source profile default)"
printf 'mode\t%s\t%s\tloud | gentle | silent\t(profile)\n' \
"$(eh_mode)" "$(eh_value_source mode profile)"
printf 'failure_spiral_threshold\t%s\t%s\t1..20\t3\n' \
"$(eh_failure_threshold)" "$(eh_value_source failure_spiral_threshold default)"
printf 'urgency_sensitivity\t%s\t%s\tlow | medium | high\tmedium\n' \
"$(eh_urgency_sensitivity)" "$(eh_value_source urgency_sensitivity default)"
printf 'judge_model\t%s\t%s\t(any model id)\tclaude-haiku-4-5-20251001\n' \
"$(eh_judge_model)" "$(eh_value_source judge_model default)"
for k in guard_test_edits guard_no_verify guard_goal_conflict \
session_baseline subagent_baseline post_compact_anchor \
enable_llm_judge enable_review_agent; do
printf '%s\t%s\t%s\ttrue | false\t(profile)\n' \
"$k" "$(eh_get_with_profile "$k")" "$(eh_value_source "$k" profile)"
done
Render the output as a markdown table with columns: Field | Current | Source | Options | Default.
Source values mean:
override — the user explicitly set CLAUDE_PLUGIN_OPTION_<field> via
/plugin config functional-emotions (or the legacy CONFIG_ env var).profile — value derived from the headline profile knob.default — value is the hard-coded static default.End with one sentence interpreting the snapshot. Examples:
balanced (default), no overrides set — the loud-mode
bundle is active."quiet, but mode is overridden to loud — banners are
on even though the bundle would have suppressed them."off; only enable_review_agent is overridden to true —
no primes fire, but the reviewer subagent stays available."Do not invent fields. If the bash block fails (e.g. CLAUDE_PLUGIN_ROOT
unset), report the failure and stop — don't substitute partial data.