sk-code-opencode

sk-code-opencode is a skill for Claude Code, OpenCode from MichelKerkmeester/skilled-agent-harness_spec-driven-loops. It costs 60 tokens per session (4,019 once invoked), scanned C, original, MIT.

A project-specific guide for developing OpenCode system files, including skills, agents, commands, plugins, configuration, and runtime connections. OpenCode is a coding-agent system.

In plain words
What is it for?
Use it when changing OpenCode skills, agents, plugins, commands, configuration, or bridge code.
Why use it?
It helps an agent follow the repository’s conventions and choose the right workflow for implementation, debugging, and verification.

Skill for Claude CodeOpenCode

Written for Claude Code and OpenCode: allowed-tools in frontmatter, but also installed under .opencode/. Also seen: mentions OpenCode.

Good fit Use it when changing OpenCode skills, agents, plugins, commands, configuration, or bridge code.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode
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.

Any agent
npx skills add MichelKerkmeester/skilled-agent-harness_spec-driven-loops --skill sk-code-opencode
Clone the repo
git clone --depth 1 https://github.com/MichelKerkmeester/skilled-agent-harness_spec-driven-loops

Made for: Claude Code, OpenCode.

Wrote 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.

agentmods badge for sk-code-opencode

README.md
[![agentmods](https://agentmods.dev/badge/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode/github.svg)](https://agentmods.dev/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode)
Your own site
<a href="https://agentmods.dev/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode"><img src="https://agentmods.dev/badge/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for sk-code-opencode

Your own site · 80×15
<a href="https://agentmods.dev/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode"><img src="https://agentmods.dev/badge/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-opencode.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 60 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,019 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00060 $0.04019
Opus 5 $0.00030 $0.02010
Sonnet 5 $0.00012 $0.00804
Haiku 4.5 $0.00006 $0.00402

Measured today against content hash dc0d78519b46, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade C, and why

sk-code-opencode scanned grade C with 1 finding 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 today.

The scan reads SKILL.md. This mod also ships 3 executable files (assets/scripts/test_verify_alignment_drift.py, assets/scripts/verify_alignment_drift.py, assets/scripts/verify_stack_folders.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Hidden instructionshighPrompt injection

Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.

<!-- Keywords: opencode, system-code, typescript, python, shell, rust, jsonc, config, mcp, skill-authoring, agent-authoring, command-authoring, hooks, alignment-verifier, surface-evidence, sk-code -->
.opencode/skills/sk-code/sk-code-opencode/SKILL.md · 192 lines

How it starts

The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.

opencode Surface — System-Code Evidence

Domain evidence and shared workflow doctrine for OpenCode system code (the .opencode/ tree: skills, agents, commands, plugins, MCP servers, config, changelogs, and runtime bridge wiring). This surface owns the implement -> debug -> verify phases through the workflow references below, then slices evidence by the detected language so a TypeScript task never pulls the Python/shell/config guides.

Detection is two-step. First, the surface trigger is work under .opencode/ (including SKILL.md, descriptors, commands, agents, plugins, MCP servers, assets, scripts, and changelogs). Second, once the OpenCode surface is selected, file extensions and local markers select the language trio: .cjs/.mjs/.js -> JavaScript, .ts/.tsx/.mts/.d.ts -> TypeScript, .py plus argparse -> Python, .sh/.bash -> shell, .rs -> Rust, .json/.jsonc/.yaml/.yml plus graph-metadata or spec-folder -> config. For Rust, when no .rs file is present, local Cargo.toml/Cargo.lock markers select it after the OpenCode surface is established; napi-rs and wasm-bindgen vocabulary are additional intent signals, not cross-project surface detectors.

1. WHEN THE HUB BUNDLES THIS

  • The task touches .opencode/ system code — a skill, agent, command, plugin, MCP server, or descriptor/config.
  • The active workflow phase needs a language standard, a language-agnostic organization pattern, a hook contract, an alignment-verification procedure, or an authoring checklist.
  • This surface owns edits, tests, and verification through the workflow references; hand off formal findings-first review to code-review and author-side quality gates to code-quality.

2. REFERENCE MAP

Language standards — after .opencode/ selects this surface, load the exact split resources for the detected language:

  • TypeScript — references/typescript/style-guide/overview-strict-and-naming.md, references/typescript/style-guide/formatting-imports-and-coexistence.md, references/typescript/quality-standards/overview-and-type-system.md, references/typescript/quality-standards/tsdoc-errors-and-async.md, references/typescript/quality-standards/tsconfig-and-modules.md, references/typescript/quick-reference/template-naming-and-types.md, references/typescript/quick-reference/imports-errors-and-tsconfig.md
  • Python — references/python/style-guide.md, references/python/quality-standards.md, references/python/quick-reference.md
  • Shell — references/shell/style-guide/overview-structure-and-naming.md, references/shell/style-guide/variables-functions-and-output.md, references/shell/quality-standards/overview-and-priority-blockers.md, references/shell/quality-standards/validation-security-and-shellcheck.md, references/shell/quick-reference/template-variables-and-loops.md, references/shell/quick-reference/functions-strings-and-checklist.md
  • Rust — references/rust/style-guide/overview-and-file-header.md, references/rust/style-guide/toolchain-and-project-structure.md, references/rust/style-guide/naming-conventions.md, references/rust/style-guide/formatting-and-imports.md, references/rust/style-guide/commenting-and-rustdoc.md, references/rust/style-guide/interop-model.md, references/rust/style-guide/interop-errors-and-parity.md, references/rust/quality-standards/overview-and-data-ownership.md, references/rust/quality-standards/modeling-collections-and-api.md, references/rust/quality-standards/docs-errors-and-async.md, references/rust/quality-standards/build-and-organization.md, references/rust/quality-standards/determinism-and-parity.md, references/rust/quick-reference/overview-and-boundary-template.md, references/rust/quick-reference/naming-ordering-and-signatures.md, references/rust/quick-reference/collections-imports-and-errors.md, references/rust/quick-reference/rustdoc-and-cargo.md, references/rust/quick-reference/determinism-parity-and-related.md
  • Config (JSON/JSONC/YAML descriptors and route assets) — references/config/style-guide.md, references/config/quality-standards.md, references/config/quick-reference.md, assets/checklists/config-checklist.md
  • JavaScript (CommonJS/ESM plugins) — references/javascript/style-guide.md, references/javascript/quality-standards/overview-modules-and-docs.md, references/javascript/quality-standards/security-testing-and-exemptions.md, references/javascript/quick-reference.md

Read the full file on GitHub · 192 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. today Changed dc0d78519b46
  2. 6d ago First seen · 192 lines · 60 tokens per session scan C aea2d7a28d1a

Subscribe to this mod's changes

sk-code-opencode is a skill published in the GitHub repository MichelKerkmeester/skilled-agent-harness_spec-driven-loops (35 stars, last pushed yesterday), licensed MIT. It adds 60 tokens to every session and 4,019 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.