cyrus / packages
CLAUDE.md888 starsChanged 53 days agoApache-2.0View on GitHub
IMPORTANT NOTE: A delegation always triggers the first agentSession on a Linear issue. An @ mention can trigger either the first or an additional agentSession on a Linear issue. For the case when it triggers an additional agentSession, the webhook MUST use the existing selected repository, as we do NOT support switching repositories within a single issue. When the first agentSession is created for a Linear issue, a repository must be selected and cached for that issue. If a repository…
1 other project uses a version of this.
Adopted by
Most recently taken up by
- cyrusagents/cyrusword for word, first seen Sep 24, 2026
Other versions1 project uses a version of this.
Word for word · 1
- cyrusagents/cyrus★ 902100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository21 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- googleSkill
- release-core-testSkill
- cyrus-setup-claude-authSkill
- cyrus-setup-endpointSkill
- cyrus-setup-githubSkill
- cyrus-setup-gitlabSkill
- cyrus-setup-launchSkill
- cyrus-setup-linearSkill
- cyrus-setup-prerequisitesSkill
- cyrus-setup-repositorySkill
- cyrus-setupSkill
- cyrus-setup-slackSkill
- debugSkill
- f1-test-driveSkill
- implementationSkill
- investigateSkill
- releaseSkill
- summarizeSkill
- verify-and-shipSkill
The file
We split this CLAUDE.md into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
No system prompt unless user explicitly uses /label-based-prompt command
- Distinctive: line 33
When this signal is received, all claudeRunners associated with this agentSession MUST be terminated.
- Distinctive: line 37
When the pendingSelection flag is set for an agentSessionCreated webhook, the subsequent agentSessionPrompted webhook will either have the …
≈ “Common” is an estimate: at least 1 in 5 of the 27,641 CLAUDE.md 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
Linear Webhook Constraints
agentSessionCreated
IMPORTANT NOTE: A delegation always triggers the first agentSession on a Linear issue. An @ mention can trigger either the first or an additional agentSession on a Linear issue. For the case when it triggers an additional agentSession, the webhook MUST use the existing selected repository, as we do NOT support switching repositories within a single issue. When the first agentSession is created for a Linear issue, a repository must be selected and cached for that issue. If a repository can not be matched based on the metadata of the Linear issue and the configured routing for the repositories, then a agentSession select signal should be sent to Linear with the configured repositories as options. In this case, a Claude runner should NOT be initialized until the subsequent agentSessionPrompted webhook is received.
An agentSessionCreated webhook has two triggers from Linear:
via @ mention:
- Skips label-based system prompt by default if (!isMentionTriggered || isLabelBasedPromptRequested))
- No system prompt unless user explicitly uses
/label-based-promptcommand - More flexible/conversational mode
via delegation:
- Uses label-based system prompt routing.
- Checks issue labels for debugger, orchestrator, or other custom prompts.
- Falls back to default system prompt.
agentSessionPrompted
An agentSessionPrompted webhook has three different handling branches:
if (agentActivity.signal === "stop"):
When this signal is received, all claudeRunners associated with this agentSession MUST be terminated. In this case, an agentSession MUST already exist.
if (this.repositoryRouter.hasPendingSelection(agentSessionId)):
When the pendingSelection flag is set for an agentSessionCreated webhook, the subsequent agentSessionPrompted webhook will either have the result of the selection or an unrelated response from the user ignoring the selection. Currently, we only use the select signal for repository selection when the agentSessionCreated webhook can not route the metadata of the Linear issue to a configured repository. In this case, a select signal is posted to Linear, which provides the user with options of the configured repositories. The user can then select a repository, which will send a agentSessionPrompted webhook where the body matches one of the options sent via the select signal, or an unrelated prompt which we should handle by just using the fallback repo (first repo configured). In both cases, a Claude runner should be initialized.
else:
For this case an agentSession MUST exist and a repository MUST already be associated with the Linear issue. The repository will be retrieved from the issue-to-repository cache - no new routing logic is performed.
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.

