agentmods.dev

abtars AGENTS.md

abtars AGENTS.md is an instructions file for Codex, OpenCode from aksika/abtars. It costs 1,386 tokens per session, scanned A, original, Apache-2.0.

Repository instructions for abtars, a TypeScript application that connects chat platforms and an agent API through a shared bridge.

In plain words
What is it for?
Use them when building, type-checking, testing, bundling, or tracing how messages move through abtars.
Why use it?
They document the main commands, startup sequence, platform adapters, and testing options needed to change the codebase safely.

Instructions file for CodexOpenCode

not rated 9 3d ago A scan 1,386 tokens original Apache-2.0
View the source file aksika/abtars
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/aksika/abtars/agents-md
Clone the repo
git clone --depth 1 https://github.com/aksika/abtars

Made for: Codex, 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 abtars AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/aksika/abtars/agents-md.svg)](https://agentmods.dev/instructions/aksika/abtars/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/aksika/abtars/agents-md"><img src="https://agentmods.dev/badge/instructions/aksika/abtars/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,386 This file is loaded in full into every session.
When invoked 1,386 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 $0.01386 $0.01386
Opus 5 $0.00693 $0.00693
Sonnet 5 $0.00277 $0.00277
Haiku 4.5 $0.00139 $0.00139

Measured 3d ago against content hash 8fc8b34300a5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

abtars 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 3d 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 · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Quick commands

npm install          # install deps
npm run build        # tsc + copy dashboard public + git build-info
npm run typecheck    # tsc --noEmit (fastest feedback)
npm test             # fast deterministic suite (unit/component + deploy logic)
npm run test:watch   # vitest (watch mode)
npm run test:integration  # integration tests (needs abmind)
npm run test:e2e         # critical end-to-end smoke/recovery tests
npm run test:extended     # integration/process/E2E tests
npm run test:all          # every Vitest test
npm run dev          # node --import tsx src/main.ts
npm run bundle       # esbuild single ESM bundle → bundle/

Run npm run typecheck before committing. It catches strict-mode errors (noUnusedLocals, noUnusedParameters, noUncheckedIndexedAccess).

Architecture

  • Entry: src/main.ts → src/bridge-app.ts → boot phases (src/boot/phase-*.ts)
  • Boot phases run sequentially in a pipeline defined in bridge-app.ts. Each phase sets exactly one field on BootCtx (src/boot/context.ts).
  • Platforms: src/platforms/{telegram,discord,agent-api}/ — adapters for chat input/output.
  • Transport: src/components/transport/ — bridges to agent CLI (tmux, ACP, or direct API).
  • Capabilities: src/capabilities/{browser,hotskills,sleep}/ — self-contained feature modules. Registry auto-generated by esbuild plugin at src/capabilities/_registry.generated.ts.
  • Components: src/components/ — core subsystems (memory, tasks, pipeline, sessions, secrets, etc.).

Model-call chokepoint: spin(spec) (#1271)

All LLM calls go through spin.spin(spec) — the single async method in src/components/spin.ts. Per-type behavior is driven by the SessionProfile registry (src/components/spin-profiles.ts); adding a new SessionType means adding a row, not a new method or branch.

  • Use spin({ type, ...spec, await: true }) for new code.
  • dispatch / dispatchAwait / sendUserToOrc are deprecated thin wrappers.
  • dispatchBackground(opts) is the entry for non-user-facing one-shots (e.g. compaction summary, system maintenance).
  • grep-invariant test (src/components/spin-invariant.test.ts) enforces: no runtime.complete( or caller-turn .sendPrompt( outside an allowlist. The allowlist is the chokepoint itself + transport machinery + the defensive fallback in openai-compat-routes.ts.

Read the full file on GitHub · 94 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. 3d ago First seen · 94 lines · 1,386 tokens per session scan A 8fc8b34300a5

Subscribe to this mod's changes

abtars AGENTS.md is an instructions file published in the GitHub repository aksika/abtars (9 stars, last pushed 3d ago), licensed Apache-2.0. It adds 1,386 tokens to every session, about $0.0069 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.

agent-securityagent-swarmagentic-aiai-agentsai-assistantai-framework
Related

Other instructions, from other repositories

Alex_Skill_Mall copilot-instructions.md

Instructions for fabioc-aloha/Alex_Skill_Mall, covering alex plugin mall: identity, mission, what the mall is, duty stack and cardinal rules.

fabioc-aloha/Alex_Skill_Mall · 1,764 tokens

Alex_Skill_Mall act-pass.instructions.md

Run the 7-step ACT pass on medium and high stakes work — Materiality first, then Hypothesise, Alternatives, Disconfirmers, Audit-priors, Severity, Commit.

fabioc-aloha/Alex_Skill_Mall · 1,586 tokens

Alex_Skill_Mall mall-maintenance-rules.instructions.md

Always-on routing for Mall maintenance work — fires the right Mall skill at the right moment. Distinguishes Mall-owned automation from out-of-scope editorial work.

fabioc-aloha/Alex_Skill_Mall · 1,288 tokens

diplomat-agent AGENTS.md

Instructions for Diplomat-ai/diplomat-agent, covering agents.md — for ai coding assistants (v0.3.0), what this repo is, architecture, key design decisions and running tests.

Diplomat-ai/diplomat-agent · 753 tokens

nextjs-skills CLAUDE.md

Instructions for mohamed-hossam1/nextjs-skills: All contribution rules — style, naming, file layout, verification — live in AGENTS.md. Read that first.

mohamed-hossam1/nextjs-skills · 69 tokens

openbotx AGENTS.md

Instructions for openbotx/openbotx, covering agents.md, build and run, setup (first time), run backend (dev mode with reload) and run frontend (dev mode, separate terminal).

openbotx/openbotx · 3,384 tokens