Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/openplanr/planr-pipelineWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/openplanr/planr-pipeline/openplanr-operate)<a href="https://agentmods.dev/rules/openplanr/planr-pipeline/openplanr-operate"><img src="https://agentmods.dev/badge/rules/openplanr/planr-pipeline/openplanr-operate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/openplanr/planr-pipeline/openplanr-operate"><img src="https://agentmods.dev/badge/rules/openplanr/planr-pipeline/openplanr-operate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00681 |
| Opus 5.5 | $0.00000 | $0.00272 |
| Sonnet 5.5 | $0.00000 | $0.00136 |
| Haiku 4.5 | $0.00000 | $0.00068 |
Grade A, and why
openplanr-operate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
OpenPlanr Operate — Cursor-native workflow
A request to run Operate means one complete Cursor-bound cycle, not a sequence of commands for the user to copy. The user does not write an orchestration prompt or run machine lifecycle commands; this rule drives the harness internally.
- Inspect with
planr operate inspect --jsonand bind tocursorfor the whole cycle. Never invoke another vendor runtime. - Execute the returned bootstrap harness when context is absent/stale. Follow
openplanr-operate-procedures/bootstrap.md: inspect the repository first and ask for one compact context review with the owner only when authority is genuinely required. Never dump a questionnaire. - Preview and start one local cycle. Follow current
handoff.next[].argvonly. - Run CEO, CTO, CPO, CMO, and COO using their generated role instructions. If Cursor has no native parallel subagents, run the same roles sequentially in Cursor; cross-vendor fallback is forbidden.
- Each role follows
openplanr-operate-procedures/advisor.md, explores the project directly under current Cursor permissions, and returns a v1.4 response withanalysisMarkdown, citedclaims, citedactions,gaps, andconflicts. Planr sends no repository evidence body and grants no permission. - Run Chair with
openplanr-operate-procedures/chair.md, finalize, and present the report usingopenplanr-operate-procedures/review.md.
Local research is automatic; connected research requires explicit per-cycle
consent. Inferred product/business context is allowed when cited and labeled;
unknowns lower confidence rather than blocking. Keep
runtimeBinding: required, crossRuntimeFallback: false, and
assurance: runtime-governed unchanged.
Use the internal harness lifecycle invisibly. The old adapter spelling is
compatibility-only. Preserve each handoff binding exactly; replay it byte-for-byte
after a transport failure. Qualified actions may create reversible provisional drafts;
do not approve, accept, apply, PLAN, SHIP, deploy, publish, spend, contact
customers, mutate credentials, or destroy data without separate authorization.
Canonical lenses: CEO (strategy-finance: Direction, business model, pricing and packaging, focus, economics, and what to stop.); CTO (technology-risk: Reliability, security, payments, privacy, data integrity, delivery risk, and blast radius.); CPO (product-activation: Actor journeys, activation, retention, friction, accessibility, and incomplete product loops.); CMO (growth-market: ICP clarity, organic demand, lifecycle coverage, proof, channel readiness, and bounded experiments.); COO (operations-customer: Human operations, billing and contracts, compliance, support load, vendors, and owner bottlenecks.); Chair (chair: Evidence reconciliation, conflict sequencing, duplicate merging, and bounded route proposals. Name any absent or not-evaluated lens as an explicit gap and never synthesize a missing lens's conclusions; consolidate only the recorded, verified board.).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 44 lines · 0 tokens per session scan A 2738487a6fef
openplanr-operate is a cursor rule published in the GitHub repository openplanr/planr-pipeline (2 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 681 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other cursor rules, from other repositories
decision-logger
Two-layer memory architecture for board meeting decisions. Manages raw transcripts (Layer 1) and approved decisions (Layer 2). Use when logging decisions after a board meeting, reviewing past decisions with /cs:decisions, or checking overdue action items with /cs:review. Invoked automatically by the board-meeting…
swamr-state-manager
Swamr State Manager — tracks task progress, manages the execution pipeline, handles retries and escalation. Maintains swamr/state.json as the single source of truth for swarm progress.
archcore-context
Archcore knowledge base context — document types, MCP tools, and conventions for working with .archcore/ documents.
sre
The local environment is optimized with rtk (Rust Token Killer). Always use the rtk prefix for shell commands (e.g., rtk npm test) to minimize token consumption.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.