run-bridge
Skill1 starsChanged 5 months agoMITView on GitHub
Start the Slack-Claude Bridge (the local process that connects Slack to Claude Code), confirm it boots and connects to Slack, and tail its logs to verify behavior. Use when you need to run the project, check that a change works end-to-end, or inspect runtime output. "The bridge" = our local process, not the AI agent reviewing code.
- Reads credentials
- Installs packages
In this repository4 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- bridge-domainSkill
- reviewing-changesSkill
- verifying-workSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this to actually run the project and observe its output — the core of verifying your own work.
- Distinctive: line 20
If they are missing, you cannot fully verify Slack behavior — say so explicitly.
- Distinctive: line 39
Run it in the background (or a separate shell) so you can inspect logs while it runs.
- Distinctive: line 49
Look for, in order:
- Distinctive: line 56
Every line prefixed with the mapping identity.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Run & verify the bridge
Use this to actually run the project and observe its output — the core of verifying your own
work. Do not claim a change works at runtime without doing this. Throughout, "the bridge"
means the local Slack-Claude Bridge process (npm run dev), never the AI assistant.
1. Preconditions
test -f .env || echo "MISSING .env — copy .env.example and ask the user to fill credentials"
npm install # if node_modules is absentThe bridge needs valid SLACK_BOT_TOKEN, SLACK_APP_TOKEN, SLACK_SIGNING_SECRET in .env.
If they are missing, you cannot fully verify Slack behavior — say so explicitly.
2. Static gate first
npm run verify # typecheck + lint + testFix failures before running. There is no point booting code that doesn't compile.
3. Start the bridge
mkdir -p logs
npm run dev # tsx watch; hot-reloads on change
# or, for a built artifact:
npm run build && npm startRun it in the background (or a separate shell) so you can inspect logs while it runs.
4. Observe output — what "working" looks like
Tail the log and confirm the expected sequence:
npm run bridge:logs # tail -f logs/bridge.logLook for, in order:
- Slack Socket Mode connection established.
- Event received for a mapped channel/thread.
- Workspace routed (correct directory for the mapping).
- Claude Code invoked inside that workspace.
- Response posted back to the correct location (channel vs thread per scope).
- Every line prefixed with the mapping identity.
5. Integration check
Where feasible, exercise the real path: send a message in a dev channel (or feed a mock event) and confirm:
mentionmappings ignore non-mention messages;automappings respond to every in-scope one.- Channel-scope replies land in the channel; thread-scope replies stay in-thread.
@claude statusreturns identity, workspace, scope, trigger, session id, messages processed, and start time.
6. Report
Quote the actual log lines / command output that prove the behavior. If you could not verify a step (e.g. no credentials, no dev channel), state which step and why — never imply success.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

