Use when about to claim work is complete, fixed, or passing — run verification commands and confirm output before any success claims; evidence before assertions always

- 因本 skill 设计缺陷导致的 bug，先用 `bensz-collect-bugs` 规范记录到 `~/.bensz-skills/bugs/`，不要直接修改用户本地已安装的 skill 源码；若有 workaround，先记 bug，再继续完成任务。

Run Cogny marketing analysis tasks — fetch scheduled tasks, analyze ad accounts via MCP, report findings

>-

Execute a hunt phase with parallel telemetry work, query logging, receipt generation, and optional wave targeting

Run all remaining phases autonomously — discuss→plan→execute per phase

Archive accumulated phase directories from completed milestones

Diagnose planning directory health and optionally repair issues

Surface the agent's assumptions about a phase approach before planning

List active THRUNT workspaces and their status

Remove a future phase from huntmap and renumber subsequent phases

abra-flexibee

acelle-mail

aci-payon

addsearch

adobe-acrobat-sign

ai21-labs

aircall

alegra

aleph-alpha

altera-payroll

alternative-payments

amadeus

amazon-cognito

amazon-ecs

amazon-eks

amazon-location-service

amazon-polly

amazon-ses

amlbot

ansible

appointo

aqilla

asana

assembla

astronaut

awesome-support

aws-s3

axis-lms

backblaze

bancontact-payconiq-company

banked

beeple

better-proposals

bexio

bigbox

bigpictureio

blackbaud

bombora

bouncer