Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.signalrooms.xyz/llms.txt

Use this file to discover all available pages before exploring further.

​
Repo rules for agents

​
⚠ Read this first

These pages are for AI agents editing files inside the private Warmr repo, not for agents using SignalRoom. If you got here because someone said “Claude, drive my morning posting run” or “GPT, check whether Lane 2 finished”. you’re in the wrong section. Go to Driving Warmr (for agents) instead. That page covers warmrctl usage, pre-flight, action commands, and recovery. Nothing here applies to live operations. Stop here if any of the following are true:
  • You are not editing files in github.com/dylanpakd-cyber/warmr (the private app repo).
  • You were not given an explicit task scope by the repo operator.
  • You are working in the canonical checkout instead of a fresh worktree (see Worktree protocol).
The cost of getting this wrong is real. Warmr is a live product running on operator hardware. Bad edits to app code, signing certs, release artifacts, or CI configuration can break posting runs in production, invalidate signed builds, or hand operators a broken update. Read these rules before any file edit. When unsure, stop and ask.
Status: Current repo-native rules. The Warmr repo files (STATUS.md, CLAUDE.md, AGENTS.md, CONTEXT.md) are the source of truth and override anything here on conflict. SignalRoom is the public product name. Warmr is the macOS app + control-plane repo that powers it. These rules govern how AI agents work inside the Warmr repo, what to read first, where to make changes, what to commit, what to leave alone.

​
Source order

Read in this order before changing anything in the Warmr repo:
  1. STATUS.md: current state, what’s shipped, what’s in flight.
  2. CLAUDE.md. Claude-specific protocol and the worktree workflow.
  3. AGENTS.md. Codex / other-agent protocol and repo rules.
  4. CONTEXT.md: the project’s domain language.
  5. Existing docs under docs/agents, docs/plans, docs/prd, docs/release, docs/adr, and docs/handoffs.
If you can’t read those, you can’t work in the Warmr repo. The public-facing docs (this site) are a curated mirror, not a substitute.

​
Work style

  • Inspect before changing. Read the file, run a freshness check, understand the surrounding code or copy.
  • Prefer small docs and code changes with clear, runnable verification.
  • Do not assume unfinished features are current. Anything labeled In progress or Planned is not shipped.
  • Use Current, In progress, and Planned labels in repo docs to mirror this discipline.
  • Keep Mintlify as a mirror of curated repo-native docs, not the source of truth.

​
Audience boundary

This section answers “I’m editing Warmr’s source, what are the rules?” If your question is one of: …then you don’t need any of the rest of this section. The pages below are for repo work specifically.

​
Continue reading