
Executor
Run a supported skill on an OPVS CRM task or board server-side. Executor fetches data, runs the skill, writes findings back and advances cards. The agent gets a job ID and checks progress without relaying bulk results through its conversation.
I built Executor for jobs where the findings belong on your CRM board. Large lists and raw API responses can fill an agent conversation before the results reach the records that need them. Executor handles that transfer on the server, so your agent can coordinate the work and check its progress.
Choose a supported skill with listSkills, then give run a task or board and an explicit source column, such as pending. The server fetches the data, runs the skill, writes findings into OPVS CRM board records and moves cards according to the result. Your agent receives a job ID and checks getStatus every 2–5 minutes.
The skill exposes 4 methods: run, getStatus, cancel and listSkills. Progress includes completed, failed and skipped counts. Cancel stops after the current batch and preserves results already written.
Use Executor when bulk findings can stay on the board. When your agent needs those findings to reason about the next step, bring the relevant data into the conversation. Install @opvs-ai/executor and ask it to enrich a board.
Features4
Run a supported skill on a task or a board with an explicit source column.
Write findings to OPVS board records and move cards according to the result.
Track completed, failed and skipped counts with a job ID.
Stop after the current batch while preserving results already written.
Videos1 video
DemoExecutor · Demo video
Demo · SpiderMediaVideos1 video
DemoExecutor · Demo video
Demo · SpiderMediaSKILL.mdthe file your agent reads
SKILL.md loads when you open this tab.
What's new3 releases · newest first · from the package's CHANGELOG
v1.0.22026-11-13Latest
- Add the approved launch-film hero and three marketplace panels. - Rewrite the store description and add four features with supported-job and status-check qualifications. - Place unchanged agent guidance and method schema in the canonical nested bundle layout; update documentation links. - Preserve the published launch-film URL. No API or execution behavior change.
v1.0.12026-11-12
The marketplace listing shows the launch film. No skill change.
v1.0.02026-06-08
- First marketplace publish of the in-tree `executor` skill (Stage 5 Wave 3). - Schema + SKILL.md carried over unchanged from the in-tree source.
Showing all 3



