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/btspoony/mstar-harnessWrote 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/btspoony/mstar-harness/mstar-cursor-plan-mode)<a href="https://agentmods.dev/rules/btspoony/mstar-harness/mstar-cursor-plan-mode"><img src="https://agentmods.dev/badge/rules/btspoony/mstar-harness/mstar-cursor-plan-mode.svg" alt="Measured on agentmods" 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 | $0.00903 | $0.00903 |
| Opus 5 | $0.00451 | $0.00451 |
| Sonnet 5 | $0.00181 | $0.00181 |
| Haiku 4.5 | $0.00090 | $0.00090 |
Grade A, and why
mstar-cursor-plan-mode 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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Morning Star × Cursor Plan mode
When the Morning Star harness plugin or /pm is active and Cursor Plan mode applies (CreatePlan / built-in plan todos):
SSOT vs CreatePlan
- SSOT for plans and status:
{HARNESS_DIR}(default.mstar/, legacy.agents/) —{PLAN_DIR}/<plan-id>-<name>.md,{HARNESS_DIR}/status.json, specs under{SPECS_DIR}when used. - CreatePlan is a session mirror only; do not treat the Cursor plan file URI as the only plan record.
CreatePlan todos (mandatory prefix)
Before any implement/code todo, CreatePlan must list these three bootstrap todos (in order) and complete them on disk:
harness-init— create harness dirs + initializestatus.jsonfrom template when missingspec-register— addplans[]row withplan_idandfilepathmirror-plan— write the SSOT main plan markdown under.mstar/plans/(or resolved{PLAN_DIR})
Implement todo done gate
Do not mark an implement todo complete until:
git commiton the authorized Working branch for that task id (no end-of-plan batch commit unless PM allowed)- matching checkbox
- [x]in the SSOT plan file status.jsonupdated when the coordination round requires it- evidence: real
git log -1 --oneline
Drift and paths
- Keep CreatePlan body and SSOT plan file aligned when scope changes.
- Use SSOT paths in Plan Path / Context Loaded, not only the Cursor plan URI.
mstar-iteration Phase 1 in Plan mode
When Plan mode is used to start a new iteration:
- CreatePlan once for a blank Phase 1 scaffold (Build-bound todos). Record that plan file path.
- Feedback loop: user gives direction/opinions only; agent explores, recommends, and edits the same plan file in place (dual-write SSOT drafts). Do not run a questionnaire loop as the main path.
- After user signals feedback-close: optional minimal deferred interview only if blocking gaps remain — still edit the same file. Never open a second CreatePlan.
- If a duplicate plan file appears: merge into the original and delete the duplicate.
- Do not dispatch Review & Edit, commit, or create the integration branch until the user clicks Build.
- Detail:
mstar-host→references/cursor-plan-mode-bridge.md§mstar-iterationPhase 1 in Plan mode.
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.
- 8d ago First seen · 62 lines · 903 tokens per session scan A fadfd7bc18c8
mstar-cursor-plan-mode is a cursor rule published in the GitHub repository btspoony/mstar-harness (58 stars, last pushed yesterday), licensed MIT. It adds 903 tokens to every session, about $0.0045 per session on Opus 5. 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 2026-08-30.
Other cursor rules, from other repositories
git-commit
A rule applied automatically when creating Git commits. It requires commit messages to omit the Co-authored-by line.
pull-request-template
A set of rules for writing GitHub pull requests, which are proposed code changes submitted for review.
commit-sponsors
Every git commit message must acknowledge BetterDesk supporters.
git-discipline
Commit, push, and branching rules. Owns ALL git behavior; other files state only their deviations.
git
Cursor rule "git" from duongductrong/cursor-kit, covering git conventions, commit messages, types, subject rules and examples.
dev-guidelines
A set of development rules for the Feishu MCP Server repository, including its workflow, issues, branches, commits, and pull requests. GitHub Flow is the practice of making changes on branches and merging them through pull requests.