warp-status-post

Files (1)

• SKILL.md

SKILL.md (first lines)

# Post status to Slack or Teams

Teams and Slack cannot pull files from a stopped agent. They can ask, and Warp can push, when the agent is running or a tick fires.

## Push now

```bash
python3 <plugin>/scripts/status_post.py --beam .warp/beam.json --out .warp/status-post.json
```

`status_post.py ?` prints `--beam` and `--out`. Quote `?` if the shell expands it.

Read `.warp/status-post.json`. It is headed `Warp | <repo> / <project>`. Post `slack` to Slack and `teams.markdown` to Teams for every messenger in `config.messenger` (`text` is the plain version). Attach `.warp/STATUS.md`, `.warp/status.json`, and `.warp/BOARD.md` with the connected Slack or Teams file tool. If the connector can only send text, paste `statusMarkdown` and say the files are in `.warp/`. If a connector is missing, append the same text to `.warp/outbox.md`.

Channels: `teamsChannel`, `slackChannel` in `.warp/config.yaml` (default `warp`, shared by all repos). If a channel is empty, post to the channel the user named.

## Pull from the channel

The one `warp-listen` listener reads the channel. This skill does not. Do not start a second reader.

When the listener drains `warp:status`, it has already written an ack (`Received warp:status. Posting the digest.`). Herald posts that ack in the channel first. Then run this skill and post the digest. The same listener acks and acts on `warp:pause`, `warp:resume`, `warp:stop`, `warp:start`, `warp:proceed <id>`, and `warp:retry <id>`. `warp:proceed` runs `proceed.py` (plan id, Jira key, or `#` number) for the one matching ticket. Reply with what it prints, including the merge sha and the Jira status after the post-merge steps.

A quiet notify setting still posts this reply. Someone asked.

From smlsr/Warp, MIT.

More from smlsr/Warp