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.
npx agentmods add rules/conlinjoe/ai-platform/70-work-itemsgit clone --depth 1 https://github.com/ConlinJoe/ai-platformWhat 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 | $0.00017 | $0.00871 |
| Opus 5 | $0.00009 | $0.00436 |
| Sonnet 5 | $0.00003 | $0.00174 |
| Haiku 4.5 | $0.00002 | $0.00087 |
Grade A, and why
70-work-items 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 2d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work-item lifecycle
Capture, don't expand scope.
Provider-neutral lifecycle:
Work item → branch → implementation → review request → approval → merge
GitHub Issues, Pull Requests, and gh are the current reference
implementation. The full policy lives in the AI Platform clone:
docs/contracts/features/work-item-lifecycle.md.
Load this rule when capturing discovered work, creating or implementing an Issue, naming a branch, or opening a Pull Request.
During an active coding session
If you discover a bug, improvement, refactor, technical debt, or feature outside the current bounded task:
- Do not fix it opportunistically.
- Do not silently expand scope.
- Determine whether it should be captured for later.
- Briefly tell the developer what was discovered.
- Offer to create a GitHub Issue (or the host's work-item equivalent).
- Continue the current task unless the discovery blocks it.
If it blocks the current task, explain why and ask how to proceed.
Do not create the Issue unless the developer authorizes it.
Issues
Issues are the backlog for work that has not begun.
Examples: bugs found during unrelated work, deferred features, technical debt, architectural follow-ups, non-blocking security or UX improvements, test-coverage gaps worth a separate task.
Keep Issues concise. Include enough context for a fresh coding agent. Canonical project docs remain authoritative.
Recommended structure; omit empty sections; do not over-document trivia:
### Problem
### Expected outcome
### Relevant context
### Acceptance criteria
Do not add Issue templates merely to enforce this shape.
Pull Requests
A PR represents implementation, not future intent. Do not create empty PRs as backlog items.
Issue → branch → implementation → verification → documentation updates where required → PR.
Reference the Issue when one exists. Describe what changed, why, verification performed, documentation/ADR/contract updates, and known limitations or follow-up work.
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.
- 2d ago First seen · 136 lines · 17 tokens per session scan A 2371ff1ac2fb
70-work-items is a cursor rule published in the GitHub repository ConlinJoe/ai-platform (2 stars, last pushed 4d ago), licensed MIT. It adds 17 tokens to every session and 871 once invoked, about $0.0001 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 2026-08-31.
Other cursor rules, from other repositories
wize-sprint-status
4-implementation: Sprint Status.
project-onboarding-rule
Automatically onboards existing projects into the AI-driven development workflow.
linear-in-review
After finishing a Linear issue, set its status to In Review (never Done).
operating-model
Engineering operating model — workflow, risk classification, scope discipline.
sprint-planning
BMAD BMM Agent: sprint-planning.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.