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 instructions/baldaworks/promptkitty/agents-mdgit clone --depth 1 https://github.com/baldaworks/promptkittyWrote 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/instructions/baldaworks/promptkitty/agents-md)<a href="https://agentmods.dev/instructions/baldaworks/promptkitty/agents-md"><img src="https://agentmods.dev/badge/instructions/baldaworks/promptkitty/agents-md.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 | $0.01407 | $0.01407 |
| Opus 5 | $0.00704 | $0.00704 |
| Sonnet 5 | $0.00281 | $0.00281 |
| Haiku 4.5 | $0.00141 | $0.00141 |
Grade A, and why
promptkitty AGENTS.md 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 4d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PromptKitty agent guidance
- Keep the root package library-first and free of CLI/process dependencies.
- Preserve the pinned PromptKit component files byte-for-byte; update them only
by changing
content/upstream.json'sref, runninggo generate ./..., and reviewing the generated content, upstream license, commit, and inventory diff. - Preserve PromptKit SPDX headers and third-party attribution.
- Assembly must resolve every declared template parameter and remain deterministic and offline.
- Run
go test ./...,go test -race ./..., the configured linter, andgovulncheckbefore completion.
PromptKitty release cycle
When asked to design, change, or run a PromptKitty release, act as a senior DevOps release maintainer and use GitHub Actions conventions only.
Grounding and safety
- Treat the repository, its workflows, the selected release tag, and live GitHub/npm state as the sources of truth. Distinguish verified facts from inferences and assumptions. Mark unknown required details as
[UNKNOWN: ...]and platform behavior that still needs checking as[VERIFY: ...]; never invent action names, versions, secrets, endpoints, paths, or release results. - Inspect before mutating. Preserve unrelated worktree changes and never include them in release commits. Do not edit pinned PromptKit component files outside the documented upstream-ref and generation flow. Preserve SPDX headers, attribution, and every bundled third-party license.
- Obtain explicit authorization for the intended SemVer version and for public publication. Do not create prereleases, alternate npm dist-tags, overwrite tags, delete releases, or unpublish packages unless the user explicitly requests that exact action.
- Keep
maindeployable. Use a descriptive branch from an up-to-datemain, conventional commits, a pull request, and squash merge. Never tag a commit that has not passed both pull-request checks and post-mergemainchecks.
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.
- 4d ago First seen · 67 lines · 1,407 tokens per session scan A 1e93fc12e1f6
promptkitty AGENTS.md is an instructions file published in the GitHub repository baldaworks/promptkitty (5 stars, last pushed 1mo ago), licensed MIT. It adds 1,407 tokens to every session, about $0.0070 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 instructions, from other repositories
gollm AGENTS.md
Instructions for teilomillet/gollm, covering agent instructions for gollm, build & test commands, build all packages, run all tests and run tests for specific package.
mcp-servercard-go CLAUDE.md
Instructions for olgasafonova/mcp-servercard-go, covering project instructions for ai agents, beads issue tracker, quick reference, rules and agent context profiles.
mcp-servercard-go AGENTS.md
Instructions for olgasafonova/mcp-servercard-go, covering agent instructions, quick reference, non-interactive shell commands, force overwrite without prompting and for recursive operations.
azure-sdk-for-go go-code.instructions.md
Instructions for Azure/azure-sdk-for-go: All code should follow the guidelines from the Azure Go SDK Guidelines. This document is a summary of the most important guidelines to follow when contributing to the Azure Go SDK.
apm python.instructions.md
Python development guidelines.
chatgpt-cli CLAUDE.md
Claude Code instructions for kardolus/chatgpt-cli, covering chatgpt-cli — release runbook, prerequisites, 1. cut the release, 2. publish the github release + binaries and 3. update the homebrew tap.