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 commands/nirelbaz/promptpit/developgit clone --depth 1 https://github.com/nirelbaz/promptpitWhat 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 | $0.00000 | $0.01589 |
| Opus 5 | $0.00000 | $0.00794 |
| Sonnet 5 | $0.00000 | $0.00318 |
| Haiku 4.5 | $0.00000 | $0.00159 |
Grade A, and why
develop scanned grade A with 1 finding 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 yesterday.
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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- Shell injection: `execSync` or `exec` with string interpolation — should use `execFile`/`execFileSync` with args array How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Full development workflow for PromptPit features. Runs design, planning, implementation, review, and simplification in one session. Use this when starting a new feature or significant change.
Phase 1: Design
-
Explore context. Read CLAUDE.md,
git log --oneline -10, and the area of code the feature touches. Understand what exists before proposing anything. -
Smart-skip assessment. Read the user's description. If it already specifies what to build, where it goes, how it works, and edge cases — skip clarifying questions and go straight to step 4. Most features from an experienced author will skip.
-
Clarifying questions (only if smart-skip didn't trigger). Ask 1-3 questions, one at a time, multiple choice preferred. Focus on approach decisions with real trade-offs, ambiguous requirements, and security model choices. Don't ask questions the code can answer.
-
Premise challenge (always run, even if skipping questions):
- Is this the right problem? Could a different framing yield a simpler solution?
- What existing code already partially solves this? Search
src/for related patterns. - What happens if we do nothing?
Take a position on each. If a premise is shaky, say so directly.
-
Propose 2-3 approaches with trade-offs. Lead with your recommendation and explain why. YAGNI: remove unnecessary features from all options.
-
Write spec to
docs/specs/YYYY-MM-DD-<topic>-design.md. Keep it concise (100-200 lines is right for most features). Commit it. -
Self-review the spec inline: placeholder scan, internal consistency, ambiguity check. Fix issues, don't re-review. Then ask the user to review — one approval gate.
Phase 2: Plan
Proceed directly after design approval. No separate skill invocation.
-
Scope check. What is the minimum set of changes that achieves the goal? If touching >8 files or introducing >2 new modules, challenge the scope before proceeding.
-
Write implementation plan to
docs/plans/YYYY-MM-DD-<topic>.mdwith this format:
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.
- yesterday First seen · 169 lines · 0 tokens per session scan A 7c49c313da98
develop is a command published in the GitHub repository nirelbaz/promptpit (6 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,589 tokens. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
fix-layout
Command "fix-layout" from OpenRaiser/PaperFit, covering /fix-layout — 启动完整 vto 排版优化, 用户入口说明, 工具调用约定, 用法 and 直接执行入口.
show-status
作用: 显示当前 PaperFit 任务的运行状态、缺陷消除进度、视觉优先级、修复计划摘要和下一步行动；这是当前默认的“摘要 / 解释”入口。.
adjust-length
作用: 尝试通过排版微调或受控语义改写逼近目标页数。它是专家快捷入口；普通自然语言如“把正文压到 8 页，尽量不要改语义”也应能触发同类任务。.
migrate-template
作用: 将论文从原模板迁移至目标模板，自动处理图表尺寸、页数、宏兼容性等问题。它是专家快捷入口；普通自然语言如“把这篇论文迁移到 CVPR 模板”也应能触发同类任务。.
repair-table
作用: 针对指定表格或当前论文中的表格问题执行修复闭环。它是专家快捷入口；普通自然语言如“修 Table 2 太挤的问题”也应能触发同类任务。.
release
You are performing a release for the mindkeg-mcp project. The user will provide the version as $ARGUMENTS (e.g., "0.2.0" or "patch" or "minor" or "major").