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.
npx skills add harshanandak/forge --skill ship-and-watchgit clone --depth 1 https://github.com/harshanandak/forgeWrote 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/harshanandak/forge/ship-and-watch)<a href="https://agentmods.dev/skills/harshanandak/forge/ship-and-watch"><img src="https://agentmods.dev/badge/skills/harshanandak/forge/ship-and-watch/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/harshanandak/forge/ship-and-watch"><img src="https://agentmods.dev/badge/skills/harshanandak/forge/ship-and-watch.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 | $0.00069 | $0.00827 |
| Opus 5 | $0.00034 | $0.00413 |
| Sonnet 5 | $0.00014 | $0.00165 |
| Haiku 4.5 | $0.00007 | $0.00083 |
Grade A, and why
ship-and-watch 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 9d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ship-and-watch
The monitor is already running. Your job is to read it, not to start it and not to replace it.
Why this exists
The single most repeated correction in this repo is an agent shipping a PR and
then hand-polling gh pr checks in a loop — burning tokens, missing the
verdict, and leaving the PR unwatched the moment the session ends.
[Forge #1 ×10]
Harsha's rule, in his words: "shepard should do everything independtly thats teh idea it shouldntbe invoked" and "whenever we ship an automatic pr monitor should be created to work on all fixes until iut gets merghe ready". A monitor you must remember to start has no value.
What is already true
Forge wakes the repo-singleton shepherd daemon automatically after a supported
session start, after every successful push, and after every successful
non-dry-run ship. Duplicate starts are clean no-ops under the O_EXCL lease.
[verified 2026-08-13 — AGENTS.md, /shepherd utility section]
So: you do not start the daemon. You confirm it is attached and read what it says.
Procedure
-
Ship or push normally —
forge ship, orforge pushon a branch that has an open PR. -
Confirm the daemon owns this PR:
forge shepherd events -
Read the verdict for your PR:
forge shepherd <pr> --pull --json -
Report the verdict in one line: what is red, what is pending, what needs a human. If nothing is red, say so explicitly and name what you checked — a silent "looks fine" reads as "did not look".
Rules
- Never hand-poll.
gh pr checksin a loop is the failure this skill replaces. The one legitimate use ofghhere is a single spot-check when the daemon reports something you want to confirm at source. [Forge #1 ×10] - Never merge. Shepherd never merges and never resolves review threads; neither do you. The human merges. [verified 2026-08-13 — AGENTS.md]
- New problems get a new PR. A finding unrelated to this branch becomes a kernel issue and its own PR, never an extra commit here. "new problems not related to pr should be menat to be fixed ina new pr." [Forge #3 ×6]
- Do not wait a fixed ten minutes. Wait for the settled state: every bot has posted a terminal result, zero unresolved threads, no check in flight. A clock is not a signal. [Forge #10 ×2]
- If the daemon is genuinely not attached, that is a bug in the substrate — file a kernel issue. Do not paper over it with a manual loop.
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.
- 9d ago First seen · 84 lines · 69 tokens per session scan A 303517524602
ship-and-watch is a skill published in the GitHub repository harshanandak/forge (4 stars, last pushed yesterday), licensed MIT. It adds 69 tokens to every session and 827 once invoked, about $0.0003 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.
Other skills, from other repositories
prowler-commit
Creates professional git commits following conventional-commits format. Trigger: When creating commits, after completing code changes, when user asks to commit.
gh-auth-isolation
Safely manage multiple GitHub identities (EMU + personal) in agent workflows.
comet-github
A routing guide for Comet-related GitHub work. It directs requests about pull requests, issues, CI failures, ideas, and fixes to the appropriate review or implementation process.
github-skill
Work with GitHub via the gh CLI — clone repositories, create/list/merge pull requests, create/list issues, and run any other gh command (API calls, workflow runs, releases, repo administration). List operations return parsed JSON.
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
nvca-chart-release
Release NVCA Operator chart changes from the native monorepo source to the vendored Helm chart. Use when updating the vendored NVCA Operator chart, changing NVCA image refs, publishing helm-nvca-operator, or validating the chart against a self-managed control plane.