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 agentmods add skills/sidcorp-co/forge/forge-releasenpx skills add SidCorp-co/forge --skill forge-releasegit clone --depth 1 https://github.com/SidCorp-co/forgeWhat 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 | $0.00089 | $0.05066 |
| Opus 5 | $0.00044 | $0.02533 |
| Sonnet 5 | $0.00018 | $0.01013 |
| Haiku 4.5 | $0.00009 | $0.00507 |
Grade A, and why
forge-release 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.
How it starts
The opening of the file, as written. The whole thing — 274 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Forge Release
The final step in the issue pipeline: released → closed. Lands the issue's own changes onto the production branch and triggers deployment. This is a lightweight skill — no codebase exploration, no review. Just land the diff, deploy, clean up.
The ISS-* branch holds this issue's changes. Release = get this issue's own changes — and only those — onto productionBranch, then deploy and close. The one non-obvious trap is called out in Step 6.
State-never-lies (VISION: state-never-lies): merged_at and the "Released" comment are promises other issues rely on — a blocks-gate/decompose child dispatches the moment merged_at is stamped, trusting the base branch now has this code. Never let branch-name equality or a push exit code stand in for verified git ancestry (see Step 4 and Step 8) — a wrongly-skipped merge or a silently-failed push must halt the release, not complete it.
Usage
/forge-release <documentId>
Tools
- forge_issues — get issue data, update status
- forge_comments — post release comment
- forge_config — get baseBranch, productionBranch, Coolify config
- forge_coolify_deploy — trigger production deploy (if configured)
- Bash — git merge, push, branch cleanup
Terminal-exit invariant
This stage may end in exactly one of these states. There is no other legal exit:
| Exit | Requires |
|---|---|
closed |
the verify-land gate returned LANDED, and the release comment is posted |
reopen |
a comment naming what is missing (gate returned NOT LANDED, or an unlandable conflict) |
waiting |
a comment naming the human decision needed |
Exiting while the issue is still at released is forbidden, including on a partial or confused run. If you cannot determine the state, post what you observed and set waiting — a silent exit at released leaves the reconciler re-dispatching this stage forever while the run holds a concurrency slot.
Re-entrancy. A re-dispatch means a prior attempt may have done part of the work. Establish what already landed before doing anything:
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.
- 2d ago First seen · 274 lines · 89 tokens per session scan A 873c8ab64ce4
forge-release is a skill published in the GitHub repository SidCorp-co/forge (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 89 tokens to every session and 5,066 once invoked, about $0.0004 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
administer-microsoft-365-tenant
Microsoft 365 tenant administration for Global Administrators. Automate M365 tenant setup, Office 365 admin tasks, Azure AD user management, Exchange Online configuration, Teams administration, and security policies. Generate PowerShell scripts for bulk operations, Conditional A…
land-and-deploy
Land and deploy workflow. (gstack).
eng-runbook
An engineering runbook — service overview, alerts table, dashboards links, common procedures with copy-pasteable commands, on-call rotation, and an incident-response checklist. Use when the brief mentions "runbook", "ops doc", "on-call guide", "SRE doc", or "运维手册".
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
gh-credit-harvest
Harvest one community PR into a release branch with authorship and credit preserved, verified green, and a warm thank-you.
release
Prepare a named version: preflight, version consistency, build/package, smoke test, checksums/notes, and release readiness. Publishing/tagging/deploy need separate authorization. Explicit-only.