seamless AGENTS.md

A set of instructions for working in the Seamless codebase, including Go style, package structure, naming, and dependency boundaries. It is specific to that repository.

In plain words
What is it for?
Making or reviewing changes in Seamless, formatting Go code, placing files in the right packages, and preserving the project’s layering rules.
Why use it?
It gives coding agents the repository’s rules so their changes fit the existing project structure and conventions.

Instructions file for CodexOpenCode

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 instructions/0spoon/seamless/agents-md
Clone the repo
git clone --depth 1 https://github.com/0spoon/seamless

Made for: Codex, OpenCode.

Per session 6,371 This file is loaded in full into every session.
When invoked 6,371 The same file — it is already loaded in full.
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 $0.06371 $0.06371
Opus 5 $0.03186 $0.03186
Sonnet 5 $0.01274 $0.01274
Haiku 4.5 $0.00637 $0.00637

Measured 2d ago against content hash bff1ba8e25cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

seamless 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.

AGENTS.md · 398 lines

How it starts

The opening of the file, as written. The whole thing — 398 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Coding conventions for AI agents working in the Seamless repository. For the project overview, structure, and commands, see CLAUDE.md.

Go style rules

General

  • Go 1.25+. No CGO. Pure-Go SQLite driver (modernc.org/sqlite).
  • Never use emojis in code or comments. Never include attribution/credit lines in commits, PRs, or code.
  • Format with gofmt. No exceptions. Prefer modern stdlib (slices, maps, strings.SplitSeq, iterators) in new code.

Package structure

Each domain package under internal/ follows a consistent layout:

  • service.go -- business logic (the public API of the package)
  • store.go -- SQLite data access
  • handler.go -- HTTP handlers (only for packages with an HTTP surface)
  • {feature}.go -- pure functions (parsers, ranking, frontmatter, etc.)
  • *_test.go -- tests alongside source

Strict layering, no circular imports. cmd/ wires everything; no package imports cmd/ and no domain package imports another's HTTP layer:

cmd/seamlessd, cmd/seam, cmd/seambench, cmd/demoseed, cmd/docsgen
  -> internal/{mcp,hooks,console}         (API surfaces)
    -> internal/{retrieve,lifecycle,gardener,files,capture,bench,demokit} (domains)
      -> internal/{store,events,llm,core,config,validate}          (foundations)

(The dependency-aware ready-queue and lease-based claiming live in internal/store/tasks*.go; there is no internal/tasks package.)

Naming

  • Interfaces: role nouns (Store, Recorder, Embedder), not IStore.
  • Constructors: New{Type}(deps) *Type (or (*Type, error) when it can fail).
  • Domain errors: package-level var Err{Condition} = errors.New(...) sentinels, checked with errors.Is.
  • Tests: Test{Unit}_{Scenario} or table-driven with t.Run.
  • IDs: ULID everywhere (github.com/oklog/ulid/v2) via core.NewID(). Never UUID.

Error handling

  • Wrap with context and %w: fmt.Errorf("pkg.Service.Method: %w", err).
  • Sentinel errors via errors.New; reserve fmt.Errorf for wrapping (with %w).
  • Either log or return an error, never both. Fatal to the operation -> return it; non-fatal (graceful degradation) -> slog.Warn/Debug and continue.

Read the full file on GitHub · 398 lines

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. 2d ago First seen · 398 lines · 6,371 tokens per session scan A bff1ba8e25cb

Subscribe to this mod's changes

seamless AGENTS.md is an instructions file published in the GitHub repository 0spoon/seamless (4 stars, last pushed 15d ago), licensed MIT. It adds 6,371 tokens to every session, about $0.0319 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.