any-coding-agent

A control guide for coding agents that do not have a dedicated integration, including Cline, Windsurf, Devin, Aider, and OpenHands. It explains how to enable agent mode and follow structured results from Agents Shipgate.

In plain words
What is it for?
Use it to run boundary checks for agent-capability changes, read JSON control results, follow permitted commands, and identify when human review is required.
Why use it?
It lets an unnamed or less-specific coding tool follow the same checks and next actions as supported integrations.

Agent

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 agents/threemoonslab/agents-shipgate/any-coding-agent
Clone the repo
git clone --depth 1 https://github.com/ThreeMoonsLab/agents-shipgate
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 905 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 $0.00905
Opus 5 $0.00000 $0.00452
Sonnet 5 $0.00000 $0.00181
Haiku 4.5 $0.00000 $0.00090

Measured 2d ago against content hash b04a51c9aa1c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

any-coding-agent 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.

docs/agents/any-coding-agent.md · 104 lines

How it starts

The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Any Coding Agent

Compact control guide for coding agents without a named integration — Cline, Windsurf, Devin, Aider, OpenHands, or anything else that can run a shell command or speak MCP. The named guides (codex.md, claude-code.md, cursor.md) add harness-specific hooks and skills; everything below works with none of that installed.

Force agent mode

Agent mode auto-enables only inside known harnesses (Claude Code exports CLAUDECODE=1, Cursor exports CURSOR_TRACE_ID). In any other harness, export the override so errors carry structured next_action / next_actions payloads instead of prose:

export AGENTS_SHIPGATE_AGENT_MODE=1

AGENTS_SHIPGATE_AGENT_MODE=0 forces it off.

The control loop

Before reporting an agent-capability change complete, run the local boundary check and parse the single stdout JSON object (shipgate.agent_boundary_result/v2):

shipgate check --agent codex --workspace . --format agent-boundary-json

--agent codex is the generic profile; use it when your harness has no named profile. Switch on control.state; follow control.next_action, control.allowed_next_commands, and control.human_review. Treat decision as diagnostic context only. Never infer a control decision from prose, Markdown, or PR comments.

check is boundary-only. If your diff adds dynamic, undeclared, or ambiguous tool capability, control.state requires verification; run the verifier:

agents-shipgate verify --workspace . --config shipgate.yaml \
  --ci-mode advisory --format json

Then read, in order: agents-shipgate-reports/agent-handoff.json first (control.state, then gate.merge_verdict), verifier.json as the authoritative control substrate, and report.json.release_decision.decision as the release gate. For committed PR refs add --base origin/main --head HEAD; make the base ref available first because verify never fetches.

Before changing host grants (MCP servers, permission rules, hooks), capture the inventory:

Read the full file on GitHub · 104 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 · 104 lines · 0 tokens per session scan A b04a51c9aa1c

Subscribe to this mod's changes

any-coding-agent is an agent published in the GitHub repository ThreeMoonsLab/agents-shipgate (87 stars, last pushed 3d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 905 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 2026-08-30.