coding-agents

Instructions for operating Codex, Claude Code, OpenCode, and other coding-agent command-line tools in persistent shared terminal sessions. A terminal session is a live command window that can remain available for follow-up work.

In plain words
What is it for?
Delegating coding tasks to interactive agents while controlling their model, reasoning effort, permissions, sandbox, and startup settings.
Why use it?
It sets rules for keeping the same session, handling prompts and approvals, and checking whether work is actually finished.

Skill for Claude CodeCodex

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 skills/vironnimo/vbot/coding-agents
Any agent
npx skills add Vironnimo/vbot --skill coding-agents
Clone the repo
git clone --depth 1 https://github.com/Vironnimo/vbot

Made for: Claude Code, Codex.

Per session 103 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,327 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 $0.00103 $0.01327
Opus 5 $0.00051 $0.00664
Sonnet 5 $0.00021 $0.00265
Haiku 4.5 $0.00010 $0.00133

Measured yesterday against content hash 6bffd8aac997, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coding-agents 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.

resources/skills/coding-agents/SKILL.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Coding Agents

Use a real interactive Terminal Session for the coding agent's entire lifecycle. The external agent, the vBot agent, and the user share the same live terminal. Terminal Sessions survive individual Runs, so an external agent may keep working after the Run that started it ends.

Non-negotiable rules

  1. Use terminal and launch the CLI's normal interactive command. Do not convert the task into a one-shot or machine-output invocation.
  2. Keep the exact terminal_id returned by start. Use that same live Terminal Session for questions, approvals, follow-up instructions, and later work instead of starting another process.
  3. Treat quiet output as an activity boundary, never as proof of completion. Use wait, then inspect status and the rendered screen.
  4. Do not disable approvals, permission checks, or sandboxes merely to avoid interaction. Answer safe, unambiguous prompts through the TUI; ask the user when the choice carries meaningful authority, risk, cost, or product intent.
  5. The user may take control of the same terminal in the WebUI. Before sending sensitive input such as an approval or menu selection, reread status and use expected_screen_revision so stale input is rejected.
  6. Never put credentials or secrets in task text. If login or trust setup appears, leave the Terminal Session available for the user and explain what needs attention.

Start or reuse

First call list. Reuse the intended live Terminal Session when one already exists. Its program-announced title, command, working directory, state, and recent screen help identify it.

For a configured CLI, call start with the interactive command, the intended workdir, any explicitly chosen model or safety arguments, and the initial instruction in text. For a registered Project, workdir: "project:<project-id>" resolves its current cwd by stable id without loading Project Context or changing Terminal ownership. start sends the text followed by Enter after the PTY is ready.

Read the full file on GitHub · 60 lines

Files

What ships with it

3 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. yesterday First seen · 60 lines · 103 tokens per session scan A 6bffd8aac997

Subscribe to this mod's changes

coding-agents is a skill published in the GitHub repository Vironnimo/vbot (1 stars, last pushed 2d ago), licensed Apache-2.0. It adds 103 tokens to every session and 1,327 once invoked, about $0.0005 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.

Related

Other skills, from other repositories

gpt-researcher

GPT Researcher is an autonomous deep research agent that conducts web and local research, producing detailed reports with citations. Use this skill when helping developers understand, extend, debug, or integrate with GPT Researcher - including adding features, understanding the architecture, working with the API…

assafelovic/gpt-researcher · 79 tokens

review-pr

Review a Dograh pull request, branch diff, or pasted patch for repo-specific security and correctness risks that are not obvious from generic FastAPI, Next.js, or Python conventions. Use when the user asks to review a PR, audit a diff, check whether changes are safe to merge, review their own changes, or asks what to…

dograh-hq/dograh · 109 tokens

gpt-researcher

Autonomous deep research from Codex via MCP.

assafelovic/gpt-researcher · 15 tokens

review-agents-md

Audit Dograh AGENTS.md files for drift against the live repo and for bad scope boundaries between parent and child docs. Use when the user asks to review existing AGENTS files, identify stale guidance, decide whether a subtree needs its own AGENTS.md, or update the AGENTS.md hierarchy under the repo root, api/, or ui/.

dograh-hq/dograh · 82 tokens

merge-pipecat-upstream

Merge the latest upstream pipecat-ai/pipecat tag into the pipecat submodule fork (dograh-hq/pipecat) and bump the dograh repo to it. Use whenever the user asks to bump, upgrade, sync, or merge pipecat, resolve pipecat merge conflicts, audit whether upstream changes break or supersede Dograh's in-fork patches, or…

dograh-hq/dograh · 102 tokens

memmachine-memory

Use when an agent or model needs durable project, user, or session context from MemMachine, needs to save information to MemMachine memory, has requests involving mem-cli, memmachine, or memmachineclient, has insufficient conversation context, or is tempted to search local files for prior context that should come from…

MemMachine/MemMachine · 96 tokens