agentmods.dev

forge-loop

forge-loop is a skill for Claude Code, Codex from harshanandak/forge. It costs 11 tokens per session (997 once invoked), scanned A, original, MIT.

A routing and resume workflow for software delivery work managed through issues and pull requests. It reads the current state, chooses the appropriate next procedure, and stops until the next wake-up when no action is due.

In plain words
What is it for?
Use it to resume delivery sessions, find ready or stale issues, monitor pull-request state, route work to the right procedure, and hand off or abandon work safely.
Why use it?
It prevents work from being started from stale context or duplicated after a pause. It also restores session settings, searches saved project rules, and handles waiting workers and existing pull requests.

Skill for Claude CodeCodex

Written for Claude Code and Codex: disable-model-invocation in frontmatter, but also agents/openai.yaml present.

not rated 0repo changed yesterday A scan 11 tokens original MIT

Good fit Use it to resume delivery sessions, find ready or stale issues, monitor pull-request state, route work to the right procedure, and hand off or abandon work safely.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/harshanandak/forge/forge-loop
View source ↗ harshanandak/forge
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.

Any agent
npx skills add harshanandak/forge --skill forge-loop
Clone the repo
git clone --depth 1 https://github.com/harshanandak/forge

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 forge-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/harshanandak/forge/forge-loop/github.svg)](https://agentmods.dev/skills/harshanandak/forge/forge-loop)
Your own site
<a href="https://agentmods.dev/skills/harshanandak/forge/forge-loop"><img src="https://agentmods.dev/badge/skills/harshanandak/forge/forge-loop/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.

agentmods 80×15 button for forge-loop

Your own site · 80×15
<a href="https://agentmods.dev/skills/harshanandak/forge/forge-loop"><img src="https://agentmods.dev/badge/skills/harshanandak/forge/forge-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 11 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 997 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00011 $0.00997
Opus 5.5 $0.00004 $0.00399
Sonnet 5.5 $0.00002 $0.00199
Haiku 4.5 $0.00001 $0.00100

Measured yesterday against content hash 0965549589fa, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

forge-loop 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 yesterday.

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.

templates/delivery/skills/forge-loop/SKILL.md · 67 lines

How it starts

The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.

forge-loop

If a worker brief says stop after push or PR open, link an open PR to the issue, release the claim, then stop; leave thread linking and CI watching to the orchestrator. With no PR, the orchestrator claims, opens and links it.

Every wake: read, route, stop. Read the chosen skill before acting.

Wake

Threaded chats: call MCP t3_thread_configuration once (no args) and set FORGE_SESSION=t3:<threadId> for every forge command, even after a harness switch. Workers you launch keep their own task session.

Only when resuming a session with existing PR work, run forge agent wait --sweep, do each step, then follow then. On abandon, an unmerged close or handback, run forge agent wait <n> --done and follow its steps and then.

Use --fallback only if a plain opening wait's link or watch step fails. Report and skip a failed --sweep or --done step.

At session start, run forge memory rules unless a hook supplied it: result.text.text is a pointer, never note text. Search memory before starting a task; use its search/show commands.

  1. Work: forge issue ready --input '{}' (ranked by priority label), forge issue stale --input '{}', then forge issue show --input '{"id":"<id>"}' for your slice.
  2. PRs: every wake runs forge pr state <n> and acts on verdict.
  3. Act only on what changed since your last push, then stop. Forge never watches: the harness wakes you.

Roles

  • Orchestrator: intake, slicing, review triage, merge, verify, re-plan.
  • Worker: one build or one fix batch on one claimed slice. Stops at push. Never merges.
  • One agent may hold both on small work.

Steps and their skill

  • 1 intake, vague intent or unresolved decisions: forge-intake
  • 2 frame, 3 slice after intake: forge-cut
  • 4 claim and workspace, 5 build, 6 check, 7 open PR: forge-build
  • 8 review triage, red PR on each wake: forge-review
  • 9 merge, 10 verify, 11 learn and re-plan; red base: forge-land
  • PR to issue is Forge's: forge issue link --input '{"id":"<slice>","relation":"delivers","url":"<pr url>"}'. It needs the held claim, so the worker links before it releases.
  • PR to agent thread: follow forge-build's opening handoff. Wakes act without re-arming event watches. After each push, run forge agent wait <n>; for t3-tool, skip its steps (already watching); otherwise do each step and follow then.

Read the full file on GitHub · 67 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. yesterday Changed · -1 lines 0965549589fa
  2. 2d ago First seen · 68 lines · 11 tokens per session scan A 34506fe02d6c

Subscribe to this mod's changes

forge-loop is a skill published in the GitHub repository harshanandak/forge (0 stars, last pushed yesterday), licensed MIT. It adds 11 tokens to every session and 997 once invoked, about $0.0000 per session on Opus 5.5. 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 2026-10-28.

Project management
Related

Other skills, from other repositories

story-readiness

Checks whether a story is ready to implement: clear acceptance criteria, no open questions, valid ADR and dependencies. Use when the user asks if a story, or the sprint's stories, are ready to start or build.

Donchitos/Claude-Code-Game-Studios · 48 tokens

recipe-create-meet-space

Create a Google Meet meeting space and share the join link.

googleworkspace/cli · 18 tokens

workthreads

SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…

specstoryai/getspecstory · 126 tokens

atmos-config

Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.

cloudposse/atmos · 31 tokens

qa-project-bootstrap

Onboard a new QA engineer to an existing codebase, or audit an existing test architecture. Produces a 30-day ramp plan: codebase orientation, framework walkthrough, test architecture audit, mentorship pairing, and first-test guidance. Use when: "QA onboarding," "new tester," "ramp up," "test architecture audit,"…

petrkindlmann/qa-skills · 124 tokens

remove

Remove a deployed framework or addon from the current workspace.

jmagly/aiwg · 12 tokens