RepoOps
Production errors, the agent session behind a change, recovery and the shared brain.
How an agent reaches it
| MCP | https://mcp.repoops.ai/mcp Built |
| Auth today | Team keys today. Sign in with PromptReports is built and off. Accepting ID agent tokens is Planned (LDG-2264). |
| ID agent token | Planned (LDG-2264) |
MCP tools
search_brain get_pre_edit_context propose_brain_entry get_incident_attribution get_prod_incidents get_agent_blast_radius spend_outcome ask_brain policy_check
Actions a person may need to approve
repoops.build_fix repoops.combine_remediation repoops.request_key_rotation fix.apply memory.lesson.promote memory.agentmd.publish
From the PromptCompute action catalog.
Workflows that use it
- Launch with proof Partly built
- Customer escalation to recovery Partly built
- Hire an agent Planned
- A trust packet a buyer can check Partly built
- Roadmap to revenue Partly built
- Cost per outcome Partly built
- Incident to policy Partly built
- Investor update with proof Partly built
- Evaluate an agent before you buy it Planned
- Ask across the family Planned
- A customer's view of what was promised Planned
- Contain an agent that went wrong Proposed
- Switch models with proof Proposed
- Answer a security questionnaire from the record Proposed
- SOC 2 and EU AI Act evidence from the record Planned
- Renewal brief with promises kept Proposed
- Release notes with proof Proposed
- What your agents did this week Proposed