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/fitchmultz/pi-codex-goal/agents-mdgit clone --depth 1 https://github.com/fitchmultz/pi-codex-goalWrote 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/fitchmultz/pi-codex-goal/agents-md)<a href="https://agentmods.dev/instructions/fitchmultz/pi-codex-goal/agents-md"><img src="https://agentmods.dev/badge/instructions/fitchmultz/pi-codex-goal/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 | $0.00637 | $0.00637 |
| Opus 5 | $0.00318 | $0.00318 |
| Sonnet 5 | $0.00127 | $0.00127 |
| Haiku 4.5 | $0.00064 | $0.00064 |
Grade A, and why
pi-codex-goal 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pi-codex-goal — agent notes
Pi extension: Codex-style /goal command and get_goal / create_goal / update_goal tools. State lives in pi session custom entries.
Local pi install policy
On this machine, the canonical active install is the global/user package pointing at this local checkout:
../../Projects/AI/pi-codex-goal
Do not leave project-local installs of this package in this repo. In particular, avoid release verification commands such as:
pi install -l npm:pi-codex-goal
pi install -l https://github.com/fitchmultz/[email protected]
Those write duplicate package entries under .pi/ for the current project, causing get_goal, create_goal, and update_goal tool-registration conflicts with the global local-checkout install. For install-path release verification, use an isolated temp project/config directory or remove the project-local entries immediately after the check. On Pi 0.84.0 or later, pass --approve for isolated project-local package install/list/non-interactive smoke commands when those commands must load .pi/settings.json. If conflicts appear, inspect pi list --approve and .pi/settings.json, then remove any project-local pi-codex-goal npm/GitHub installs so only the global local-checkout package remains active.
Verify before finishing
npm run verify
Runs tsc --noEmit, the platform-smoke harness checks, and the full Node test suite (test/*.test.ts).
For release-sensitive changes, also use the local Crabbox platform gate documented in docs/platform-smoke.md:
npm run check:platform-smoke
npm run smoke:platform:all
smoke:platform:all runs smoke:platform:doctor before any target suite starts.
The required gate runs the full suite plus a real model-backed goal-tool smoke on macOS, Ubuntu Linux, and native Windows. The default smoke model is zai/glm-5.2; override with PLATFORM_SMOKE_MODEL when needed.
Layout
| Area | Modules |
|---|---|
| Wiring | src/index.ts, goal-runtime-controller.ts |
| User / model API | commands.ts, tools.ts, prompts.ts, format.ts, clipboard.ts, prompts/create-goal.md |
| Runtime events | goal-runtime-event-handlers.ts, goal-runtime-*-handlers.ts |
| Transitions | goal-transition.ts, goal-transition-effects.ts, goal-state-controller.ts |
| Stale continuations | stale-queued-work-*.ts |
| Recovery | recovery*.ts |
| Domain | state.ts, types.ts, goal-persistence.ts |
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 · 54 lines · 637 tokens per session scan A 49f7b47ef5f5
pi-codex-goal AGENTS.md is an instructions file published in the GitHub repository fitchmultz/pi-codex-goal (181 stars, last pushed 28d ago), licensed MIT. It adds 637 tokens to every session, about $0.0032 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
pi-agent-browser-native AGENTS.md
Instructions for fitchmultz/pi-agent-browser-native, covering pi-agent-browser-native, product-specific guidance, documentation placement, upstream capability baseline and command reference and tool result categories.
pi-commandcode-provider AGENTS.md
Instructions for patlux/pi-commandcode-provider, a project described as: pi custom provider for the Command Code API — 18 models including Claude, GPT, DeepSeek, Kimi, GLM, MiniMax, and Qwen.
pi-fancy-footer AGENTS.md
Instructions for mavam/pi-fancy-footer, covering pi-fancy-footer, setup, development and release engineering.
pi-subagents AGENTS.md
AGENTS.md instructions for tintinweb/pi-subagents, covering development rules, conversational style, code quality, documentation and commands.
pi-anthropic-auth AGENTS.md
Instructions for gotgenes/pi-anthropic-auth, covering agents guide: pi-anthropic-auth, project, primary goal, current status and principles.
supi CLAUDE.md
Instructions for mrclrchtr/supi, covering claude.md, development status, pi docs-first rule, documentation expectations and package layout convention.