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/promptdriven/pdd/agents-mdgit clone --depth 1 https://github.com/promptdriven/pddWhat 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 | $0.00604 | $0.00604 |
| Opus 5 | $0.00302 | $0.00302 |
| Sonnet 5 | $0.00121 | $0.00121 |
| Haiku 4.5 | $0.00060 | $0.00060 |
Grade A, and why
pdd 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 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 — 25 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
Core CLI logic lives in pdd/, where each module aligns with a prompt template in prompts/ (for example, pdd/code_generator.py pairs with prompts/code_generator_python.prompt). Generated examples drop into context/, CSV data into data/, and packaged artifacts into dist/. Public examples live under examples/, and central documentation, diagrams, and onboarding notes reside in docs/.
Build, Test, and Development Commands
Activate the Conda environment once per session with conda activate pdd; all Makefile targets assume interpreter and deps from that env.
make testruns the staging pytest suite defined intests/.make coverageexecutes pytest with coverage; review the report before large merges.make regressionormake sync-regressiontriggers the longer regression harnesses (optionally passTEST_NUM=42).make lintwrapspylintchecks; fix diagnostics instead of suppressing them.make buildandmake installcreate and install the CLI wheel locally for smoke checks.
Coding Style & Naming Conventions
Use Python 3.12+, four-space indentation, and type annotations for public functions. Match module and file names to their prompt identifiers (snake_case), keep classes in PascalCase, and reserve UPPER_SNAKE_CASE for constants. Favor small, composable functions with docstrings summarizing side effects. Run make lint or pylint pdd/<module>.py before submitting to catch style regressions.
Testing Guidelines
All automated tests live in tests/test_*.py and follow the pytest discovery rules pinned in pytest.ini. Name new tests test_<module>_<behavior> and colocate fixtures in the same file unless shared broadly. Use pytest -k sync_main for targeted runs while iterating, then make coverage to confirm branch coverage holds. Regression shells (tests/regression.sh, tests/sync_regression.sh) expect API access—flag them as real with the provided marker when applicable.
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 · 25 lines · 604 tokens per session scan A 01a432340a83
pdd AGENTS.md is an instructions file published in the GitHub repository promptdriven/pdd (861 stars, last pushed 8d ago), licensed MIT. It adds 604 tokens to every session, about $0.0030 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 instructions, from other repositories
ai-interview-kit AGENTS.md
Instructions for CyannSHI/ai-interview-kit, covering ai外呼 prompt 工程化工具包, 可用技能 and 开场引导.
dev-challenge CLAUDE.md
Instructions for micheltlutz/dev-challenge, covering claude.md, the one rule to internalise, slash commands, subagents and skills.
mint AGENTS.md
Instructions for 3li7alaki/mint, covering mint: atomic completion ledger contract, one unit, one claim, state and worktrees, attempts and provenance and verification and reviews.
AgentX ai.instructions.md
AI and ML specific coding instructions for building AI agents, LLM integrations, and intelligent workflows.
LineCodePro CLAUDE.md
Claude Code instructions for LangLang03/LineCodePro, covering claude.md, project, module layout, commands and architecture.
vscode-copilot-chat model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.