Skip to content

Add sre-diagnose Claude Code skill for subscription-mode runs - #211

Closed
pintxxo wants to merge 1 commit into
fuzzylabs:mainfrom
pintxxo:feature/sre-diagnose-skill
Closed

Add sre-diagnose Claude Code skill for subscription-mode runs#211
pintxxo wants to merge 1 commit into
fuzzylabs:mainfrom
pintxxo:feature/sre-diagnose-skill

Conversation

@pintxxo

@pintxxo pintxxo commented Sep 9, 2026

Copy link
Copy Markdown

Adds a project skill that reproduces the agent workflow from core/agent.py without an Anthropic API key. Claude Code acts as the agent loop, the local cw-mcp-server replaces the boto3 CloudWatch tool, the Slack connector replaces the Slack MCP sidecar, and a local checkout or the gh CLI replaces the GitHub MCP server.

The skill mirrors system_prompt.txt and diagnosis_prompt.txt: open a Slack thread first, query logs, read code only when logs exist, reply in the same thread, and return an ErrorDiagnosis JSON. It adds a Logs Insights fallback for log groups that do not use the log_processed envelope, and uses explicit start/end windows because Insights rejects relative @timestamp filters.

defaults.json holds per-user values (log group, Slack channel, AWS profile and region) and is overridden by command-line flags.

Write a short description which explains what this pull request does and, briefly, how.

What

Why

How

Extra

If new dependencies are introduced to the project, please list them here:

  • new dependency

Checklist

Please ensure you have done the following:

  • I have run application tests ensuring nothing has broken.
  • I have updated the documentation if required.
  • I have added tests which cover my changes.

Type of change

Make sure to update label on right hand panel.

MacOS tests

To trigger the CI to run on a macOS backed workflow, add the macos-ci-test label to the pull request (PR).

Our advice is to only run this workflow when testing the compatability between operating systems for a change that you've made, e.g., adding a new dependency to the virtual environment.

Note: This can take up to 5 minutes to run. This workflow costs x10 more than a Linux-based workflow, use at discretion.

Adds a project skill that reproduces the agent workflow from
core/agent.py without an Anthropic API key. Claude Code acts as the
agent loop, the local cw-mcp-server replaces the boto3 CloudWatch
tool, the Slack connector replaces the Slack MCP sidecar, and a local
checkout or the gh CLI replaces the GitHub MCP server.

The skill mirrors system_prompt.txt and diagnosis_prompt.txt: open a
Slack thread first, query logs, read code only when logs exist, reply
in the same thread, and return an ErrorDiagnosis JSON. It adds a Logs
Insights fallback for log groups that do not use the log_processed
envelope, and uses explicit start/end windows because Insights rejects
relative @timestamp filters.

defaults.json holds per-user values (log group, Slack channel, AWS
profile and region) and is overridden by command-line flags.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
@pintxxo

pintxxo commented Sep 9, 2026

Copy link
Copy Markdown
Author

Opened against upstream by mistake; this skill is for our fork only. Closing.

@pintxxo pintxxo closed this Sep 9, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant