Notebook session ended
This session ended after eight hours.
Sign in again to continue from where you left off.
Notebook gate offline
The sign-in service is not responding right now.
Wait a minute and try again, or read what is already public from the hub.
Foundational walk through the petrova project. Read top to bottom if you're new; no prior context assumed.
More about this project on the Petrova hub.
Use when: deciding whether content belongs in petrova-hq or a consumer repo; explaining why petrova-hq has no app code or test runner; tracing how a methodology change reaches every governed repo
Use when: editing a prompt or template and seeing 'modified content' in petrova-hq; deciding whether to commit in the parent or in a submodule; rolling out a methodology change to consumers via pointer bump
Use when: auditing a repo for governance compliance; deciding whether a proposed change violates a cross-cutting rule; writing a refusal_condition for a new subagent
Use when: deciding whether a phase is ready to close; reading a MILESTONES.md status block; explaining why open / next / closed are the only valid transitions
Use when: running 00-bootstrap.md against a fresh consumer repo; auditing why a generated CLAUDE.md contains a particular invariant; explaining why MR-15 makes Q1-Q7 schema-bound
Use when: starting or ending a phase via Claude Code; writing a phase-open or phase-close decision doc by hand; explaining why phase prompts are deliberately interactive
Use when: running 03-verification-round.md inline at phase close; deciding which probe shape (A-F) applies to the phase being closed; explaining why zero findings is a failure mode
Use when: a PR feels off but you cannot articulate why; a doc just landed that contradicts another; deciding whether a change is scope creep or a real new milestone
Use when: revising a closed verdict; writing a decision doc that overturns an earlier one; auditing a repo for MR-7 compliance
Use when: adding a new doc category to a consumer repo; debugging a docs-invariants CI failure; auditing whether the rank-graph DAG still holds
Use when: adding a new subagent role; wiring spawns_on triggers for an existing role; explaining MR-6 (read AGENTS.xml, never improvise delegation)
Use when: reading or editing a per-repo state file; explaining why probes auto-demote but never auto-promote; auditing the difference between intent (contract) and observation (state)
Use when: debugging why the CASA Petrova zone is stale; rotating HUBBLE_INGEST_TOKEN; explaining the daily 07:00 UTC snapshot path
Use when: wiring a consumer repo to one of the five integration surfaces; deciding which prompt to paste for an integration mission; auditing whether a wire-up adheres to the gated-taskset discipline
Use when: choosing the right petrova-* skill for an operator question; auditing why a write skill refused to proceed; mapping a verb invocation back to its containing skill
Use when: writing the petrova homepage hero; explaining petrova to a founder shipping with Claude Code
Use when: qualifying inbound interest; deciding whether a project is ready to adopt petrova
Use when: writing the homepage hero or a tweet; pitching petrova to another developer in under a minute
Use when: telling the story of how petrova got here; explaining what shipped this phase and what's next
Use when: responding to 'isn't this just Nx or Turborepo?'; comparing petrova against hand-rolled scripts or a wiki
Use when: framing why repo-level agent context is a venture-shaped category; answering 'why now' for AI-agent orchestration in software organizations
Use when: sizing the addressable market for repo-level AI-agent tooling; answering 'who pays for this' in a partner conversation
Use when: answering 'what stops Cursor or GitHub from building this in a weekend'; explaining why the defensibility lives in the discipline, not the file format
Use when: answering 'who is building this' in a first-meeting; framing solo-builder velocity as evidence, not a risk