sage-code

A code-generation command that reads a project guide, a specification, and behavior.md, a file describing expected scenarios and outcomes.

In plain words
What is it for?
Use it to generate code and corresponding tests from completed specifications and behavior scenarios. TDD, or test-driven development, means using tests to define expected behavior; this applies that idea at the prompt level.
Why use it?
It makes the generated code follow the project's stated conventions and acceptance criteria. It stops first if behavior.md still contains unresolved questions.

Command

Install

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.

agentmods
npx agentmods add commands/gustavobarbosab/sage/sage-code
Clone the repo
git clone --depth 1 https://github.com/gustavobarbosab/sage
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 395 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00395
Opus 5 $0.00000 $0.00198
Sonnet 5 $0.00000 $0.00079
Haiku 4.5 $0.00000 $0.00040

Measured yesterday against content hash 7fdd518c06c2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sage-code 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.

commands/sage-code.md · 56 lines

What it actually says

/sage-code

Use this prompt to generate production-ready code from spec.md + behavior.md.

The AI has full context at this point: harness, spec, and a behavioral contract. The code is generated to satisfy the behavior scenarios — TDD's spirit applied at the prompt level.


Prompt

You are SAGE, a spec-first AI development assistant.

Read:
- The harness file (.sage/harness.md or in project knowledge)
- spec.md
- behavior.md

BEFORE GENERATING CODE — check behavior.md for unresolved open questions.
If there are any `- [ ]` items, STOP and remind me to resolve them first.

If all open questions are resolved, generate production-ready code that:

1. Follows the harness conventions EXACTLY — stack, naming patterns, architecture, restrictions
2. Implements every acceptance criterion from spec.md
3. Satisfies every scenario in behavior.md with corresponding test code
4. Includes previews/examples where the harness requires them
5. Respects every "Do NOT" item from spec.md

Format the output as multiple files. For each file, start with:

// FILE: <relative/path/to/File.kt>

Then the file contents.

Do not include explanations between files unless I ask for them.
Do not deviate from the architecture defined in the harness.

Workflow

  1. Verify behavior.md has no unchecked open questions
  2. Run this prompt
  3. Review the output against spec.md acceptance criteria — point by point
  4. If something needs changing, use sage-update.md for precise feedback

Tips

  • Review against the spec, not against your gut feel
  • Precise feedback ("scenario X is missing the assertion for Y") works better than vague feedback ("this doesn't look right")
  • If the AI deviates from the harness, that's a signal your harness needs more detail
Changes

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.

  1. yesterday First seen · 56 lines · 0 tokens per session scan A 7fdd518c06c2

Subscribe to this mod's changes

sage-code is a command published in the GitHub repository gustavobarbosab/sage (2 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 395 tokens. 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.