Ankh Development

Ankh Development is a skill for Claude Code, Codex from Abruptive/Ankh.md. It costs 29 tokens per session (1,849 once invoked), scanned A, original, MIT.

A development guide for ankh, a Hermes-based coding agent that applies separate .agent/ settings in each folder. It documents installation, version upgrades, patching, and end-to-end checks.

In plain words
What is it for?
Installing ankh, changing its pinned Hermes version, rebuilding it, fixing failed patches, and running repository end-to-end tests.
Why use it?
It gives maintainers a consistent way to update the bundled agent and diagnose patch failures. It also records the commands used to confirm an upgrade works.

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/abruptive/ankh.md/ankh-development
Any agent
npx skills add Abruptive/Ankh.md --skill ankh-development
Clone the repo
git clone --depth 1 https://github.com/Abruptive/Ankh.md

Made for: Claude Code, Codex.

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 Ankh Development

README.md
[![agentmods](https://agentmods.dev/badge/skills/abruptive/ankh.md/ankh-development.svg)](https://agentmods.dev/skills/abruptive/ankh.md/ankh-development)
Your own site
<a href="https://agentmods.dev/skills/abruptive/ankh.md/ankh-development"><img src="https://agentmods.dev/badge/skills/abruptive/ankh.md/ankh-development.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,849 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 $0.00029 $0.01849
Opus 5 $0.00015 $0.00924
Sonnet 5 $0.00006 $0.00370
Haiku 4.5 $0.00003 $0.00185

Measured 4d ago against content hash a79690d027d9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Ankh Development 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 4d 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.

.agent/skills/code/ankh-development/SKILL.md · 180 lines

How it starts

The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ankh Development Skill

Techniques and shortcuts for developing ankh. For global repo facts and preferences, see AGENTS.md.

Quick Start

bun install
bun bootstrap

Add ~/.agent/extensions/ankh/bin to PATH. Run hermes ankh or ankh for usage.


Version Upgrade

1. Update pinned version

Edit src/vendor.json (source of truth):

"source": {
  "type": "git",
  "url": "https://github.com/NousResearch/hermes-agent",
  "branch": "v2026.3.12",
  "version": "v2026.3.12"
}
  • Set version and branch to the desired Hermes tag (e.g. v2026.3.13).

2. Fresh clone and reapply patches

Patches are applied against the cloned Hermes source. After a version bump, patches may fail:

# Force a clean cache for the new version
FRESH_CACHE=1 bash src/scripts/build/install.sh

If a patch fails, install.sh will exit with an error. Fix the patch in src/patches/core/<name>.patch and retry.

3. Patch resolution order

The apply order is defined in src/scripts/build/install.sh (the for patch_name in ... loop). Post-patch hooks: directory runs identity.py; banner runs banner.py.

4. Verify after upgrade

bun bootstrap
bun run test:e2e:repo
bun run test:e2e:tarball

Build, Deploy, Test

Command Purpose
bun run install Clone Hermes, apply patches → .build/cache/<version>/
bun run build Create venv, bundle patched source → .build/
bun run deploy Copy to ~/.agent/extensions/ankh/runtime, create wrappers in ~/.agent/extensions/ankh/bin
bun run test Validate build, deploy, scoping, uninstall flows
bun bootstrap install → build → deploy
bun run reset Remove .build

Shortcuts

  • Full cycle: bun bootstrap
  • Re-patch only (no re-clone): bun run install (uses existing cache if hash matches)
  • Fresh patch run: FRESH_CACHE=1 bun run install
  • Override paths (for CI or isolation): CACHE_DIR, BUILD_DIR, HOME, VERSION

Read the full file on GitHub · 180 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. 4d ago First seen · 180 lines · 29 tokens per session scan A a79690d027d9

Subscribe to this mod's changes

Ankh Development is a skill published in the GitHub repository Abruptive/Ankh.md (74 stars, last pushed 5mo ago), licensed MIT. It adds 29 tokens to every session and 1,849 once invoked, about $0.0001 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 2026-08-30.