Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/bnet47/codexiconnpx agentmods add skills/bnet47/codexicon/shipWrote 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.
[](https://agentmods.dev/skills/bnet47/codexicon/ship)<a href="https://agentmods.dev/skills/bnet47/codexicon/ship"><img src="https://agentmods.dev/badge/skills/bnet47/codexicon/ship/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.
<a href="https://agentmods.dev/skills/bnet47/codexicon/ship"><img src="https://agentmods.dev/badge/skills/bnet47/codexicon/ship.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01172 |
| Opus 5.5 | $0.00000 | $0.00469 |
| Sonnet 5.5 | $0.00000 | $0.00234 |
| Haiku 4.5 | $0.00000 | $0.00117 |
Grade A, and why
ship 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ship
Determine the authorization ceiling from the request before acting:
- Commit only: verify and create the commit, then stop.
- Push: verify, commit when needed, and push the current non-protected branch; do not open a PR.
- Open a PR / ship: verify, commit, push, and open a draft PR.
None of these requests authorizes deployment or unrelated cleanup.
Verification tiers and ownership
Focused checks belong to the current Build writer, and Build completion belongs to the primary Build writer. Those receipts may be reviewed during Ship, but they do not satisfy Ship. Ship is human-owned and begins only after explicit authorization; it is the only tier that may perform user-checkout Git, tracked/history, release, publication, deployment, or external-write work.
Ship evidence must be fresh within the selected profile's Ship window and must follow the latest relevant write. Safe read-only inspection preserves valid evidence, but protected credential paths are never opened. A mutation, unsafe or unknown command, malformed state, or failed tracked/history enumeration invalidates the affected evidence and requires the relevant check again.
Acceptance by authorization ceiling
The requested Ship ceiling determines the completion gate:
- Commit-only: require the full lint, test, filesystem-security, and tracked/history checks plus the change-set audit and review. A commit-only request is accepted after the authorized commit; it does not require release identity, publication evidence, a remote push, merge, or deployment, and the workflow must not pressure the user to perform any of them.
- Publish, merge, or deploy: require explicit authority for that action in addition to the full lint/test/security and tracked/history checks. Release identity and publication evidence are required before the authorized action; merge or deployment must not be inferred from a commit or push request.
Tier-aware Ship evidence records verification_tier: "ship", the selected
profile, source/contract/check identities, and the configured freshness window,
expires_at, and responsible owner. The human Ship owner reruns the affected
checks after expiry or invalidation; these fields document responsibility and do
not create a runtime.
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.
- 1mo ago Changed · +56 lines 1138191709d6
- 1mo ago Changed · +1 lines f8cc51ec0792
- 1mo ago Changed 34141b648756
- 1mo ago First seen · 65 lines · 0 tokens per session scan A 0350ee11758a
ship is a skill published in the GitHub repository bnet47/codexicon (5 stars, last pushed 20d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,172 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 2026-08-31.
Other skills, from other repositories
finishing-a-branch
Wrap up finished code work safely: confirm the tests pass, then let the Commander choose to merge it, open a pull request, or keep it for later. Nothing is deleted or overwritten without the Commander's clear go-ahead.
publish-site
Put a finished website online on the Commander's own hosting account (GitHub Pages, Cloudflare Pages or Netlify): preview it, publish a saved version, check the live address really works, and keep a way to roll back.
oma-scm
Manage Git branches, merges, conflicts, commits, and release baselines. Use for repository history and change-management operations.
publish-harness
Publish a generated harness to npm — runs the smoke test, signs the witness manifest, and dispatches npm publish --provenance from your tagged release.
changelog-writing
Write the changelog between two versions: gather what changed, group it by what users notice, say each change in one plain sentence, credit contributors, and put breaking changes and their fixes first.
issue-triage
Sort a backlog of issues and pull requests: label each by type and severity, link duplicates, try the bug reports, ask for the one missing detail, and turn the accepted ones into ready-to-work briefs.