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/tetracoralla/decision-table/agents-mdgit clone --depth 1 https://github.com/tetracoralla/decision-tableWrote 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/tetracoralla/decision-table/agents-md)<a href="https://agentmods.dev/instructions/tetracoralla/decision-table/agents-md"><img src="https://agentmods.dev/badge/instructions/tetracoralla/decision-table/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 5.1 | $0.00500 | $0.00500 |
| Opus 5 | $0.00250 | $0.00250 |
| Sonnet 5 | $0.00100 | $0.00100 |
| Haiku 4.5 | $0.00050 | $0.00050 |
Grade A, and why
decision-table 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 5d 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Decision Table repository contract
Product boundary
This repository owns a deterministic Agent-facing decision interface. It does not own workflows, business process orchestration, policy authoring by an LLM, general expression evaluation, optimization, or a visual rule editor.
Read docs/PRODUCT_MODEL.md and docs/REVIEW_CONTRACT.md before changing or
reviewing the product.
A plain owner request to review, audit, 审核, or 复核 automatically invokes the
complete review contract in read-only mode unless fixes are also requested.
Treat it as the minimum scope, not a ceiling, and finish with tools-dev workspace escalations for shared contracts, installation, or resource risks;
do not ask the owner to supply a separate checklist.
The canonical model is docs/PRODUCT_MODEL.md; the executable Zod schemas in
src/model/schemas.ts are the transport and runtime authority. Keep examples,
CLI, MCP, and the Codex Skill on that one model.
Invariants
- Never evaluate source strings, JavaScript, regular expressions, or an opaque expression object. Conditions use the tagged canonical IR only.
- Missing data is
UNKNOWN, notfalseornull. Preserve three-valued semantics throughall,any, andnot. - Decimal facts are canonical strings. Do not silently accept binary floating
point for
decimalinputs. - Reject unknown request, ruleset, condition, and fact fields. A typo must not silently change a decision.
- Every result identifies the exact ruleset id, version, and deterministic fingerprint that produced it. Version and effective-window failures are ordinary structured statuses.
- Static analysis must only report mechanically proven overlap or shadowing. Do not imply full SAT/SMT completeness.
- MCP tools are read-only and idempotent. They accept inline JSON only; do not add ambient filesystem authority without a separately granted root and path escape tests.
- Keep the ordinary result envelope below the declared response limit and all request-wide node, depth, input, and rule limits cumulative.
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.
- 5d ago First seen · 50 lines · 500 tokens per session scan A 7debabd45830
decision-table AGENTS.md is an instructions file published in the GitHub repository tetracoralla/decision-table (0 stars, last pushed 6d ago), licensed Apache-2.0. It adds 500 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
mcp-steroid CLAUDE.md
Claude Code instructions for jonnyzzz/mcp-steroid, covering claude.md, agents.md, design philosophy, recursive context lookup (do this before sub-folder work), sub-folder guides and must do.
autovault CLAUDE.md
Claude Code instructions for autoworks-ai/autovault, covering claude.md, project, common commands, architecture and request flow.
anywidget-mcp AGENTS.md
Instructions for peter-gy/anywidget-mcp, covering anywidget-mcp, commands, architecture, dependency rule and state invariant.
codegraph AGENTS.md
AGENTS.md instructions for lzehrung/codegraph, covering agent directives, general and path, cache, and review safety.
lianhuanhua-skills AGENTS.md
Instructions for littlewindy123/lianhuanhua-skills, covering agents.md, project goal, architecture rules, commands and before committing.
ai-plugin GEMINI.md
Gemini CLI instructions for PostHog/ai-plugin, covering posthog extension, available tool categories and guidelines.