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/painhardcore/pstack/agents-mdgit clone --depth 1 https://github.com/painhardcore/pstackWhat 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 | $0.00493 | $0.00493 |
| Opus 5 | $0.00246 | $0.00246 |
| Sonnet 5 | $0.00099 | $0.00099 |
| Haiku 4.5 | $0.00049 | $0.00049 |
Grade A, and why
pstack 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 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.
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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pstack instructions for Codex
Codex loads this repository through .codex-plugin/plugin.json. Type $ and select a pstack: skill for explicit invocation. Codex can also select a skill implicitly from its description.
Use pstack:poteto-mode as the entry point for non-trivial engineering work.
Direct skills
poteto-moderoutes work to playbooks and principles.how,why,recall, andblast-radiusestablish context.architect,arena,swarm, andinterrogateshape or review work.teach,tdd,no-comments, andtypescript-best-practiceshandle focused workflows.figure-it-outandshow-me-your-workhandle large or auditable work.create-verification-skillandmaintain-verification-skillbuild project verification.technical-writing,unslop, andbroshape communication.
The 21 principle-* skills are loaded by direct skills when their rules affect a decision.
Portable behavior
Use Codex's native planning, collaboration, file, terminal, wait, and automation tools. Inherit the parent model unless a task explicitly needs another available model. If the user prohibits subagents, run the workflow inline and keep the skipped delegation step with its reason.
Repository-local skills belong under .agents/skills/.
Anything that changes external state requires explicit authority. This includes pushes, pull requests, merges, deployments, messages, ticket updates, and remote deletion.
Playbooks
poteto-mode ships 19 playbooks:
- investigation, bug-fix, perf-issue, hillclimb, runtime-forensics, trace-forensics
- feature, refactoring, prototype, visual-parity, authoring-a-skill, eval
- babysit, autonomous-run, session-pickup, pause-safely
- multi-phase-plan, worktree-cleanup, opening-a-pr
Cursor-only skills and playbooks are not included. Use the upstream Cursor plugin for those workflows.
Repository checks
Run the complete public release gate before reporting repository changes as complete:
make check
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 · 47 lines · 493 tokens per session scan A bc5715fb66e0
pstack AGENTS.md is an instructions file published in the GitHub repository painhardcore/pstack (1 stars, last pushed 5d ago), licensed MIT. It adds 493 tokens to every session, about $0.0025 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
vibes-plug AGENTS.md
Instructions for roedyrustam/vibes-plug, covering core rule for vibes-plug agents, mandatory: skill orchestration update, mandatory: multi-agent orchestration & swarm cooperation, swarm execution topologies (2026 edition) and swarm director protocols.
vibes-plug CLAUDE.md
Instructions for roedyrustam/vibes-plug, covering how this plugin works, core rules, 1. deep reasoning before action, 2. skill orchestration protocol and 3. bilingual ecosystem.
coding-agents-skills AGENTS.md
Instructions for svngoku/coding-agents-skills, covering coding-agents-skills — agent instructions, repository overview, repository structure, skill conventions and 1. yaml frontmatter (required).
codealive-skills AGENTS.md
Instructions for CodeAlive-AI/codealive-skills, covering codealive-skills — notes for ai coding agents, what the skill is not, architecture in one sentence, releasing and writing the skill description.
copilot-prompt AGENTS.md
AGENTS.md instructions for jim60105/copilot-prompt: Operate with native-level English fluency while also communicating in 正體中文 zh-tw. Perform all internal reasoning in English like professionals. Execute all tasks in English. Write all commit messages in English. Write all code comments in English. Produce all…
opentikz CLAUDE.md
Claude Code instructions for opentikz/opentikz, covering opentikz — project context for claude code, what this project is, target user, explicit non-goals (do not build these) and the three-layer content model.