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 firatcand/forge --skill drivegit clone --depth 1 https://github.com/firatcand/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/firatcand/forge/drive)<a href="https://agentmods.dev/skills/firatcand/forge/drive"><img src="https://agentmods.dev/badge/skills/firatcand/forge/drive.svg" alt="Measured on agentmods" 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 | $0.00056 | $0.03811 |
| Opus 5 | $0.00028 | $0.01906 |
| Sonnet 5 | $0.00011 | $0.00762 |
| Haiku 4.5 | $0.00006 | $0.00381 |
Grade A, and why
drive 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 7d 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 — 324 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/drive
/drive <ticket-id> drives one tracker ticket as far toward "merged + Done"
as a single invocation can — claim, plan, implement, verify, cross-review, ship,
merge, mark Done, wrap up — pausing only for architectural decisions or when a
gate cannot be satisfied. It is the per-ticket recipe that codifies the proven
v0.4/v0.5 hardening-loop gauntlet (the exact loop this repo's own batches were
delivered by, by hand) into a reusable, settings-configurable forge skill that
orchestrates the EXISTING per-task skills.
This is a per-turn recipe — NOT a loop (ADR)
/drive does all it can in one invocation and then RETURNS. It MUST NOT
self-loop: no sleep, no watch, no while true, no --follow, no
gh pr checks --watch. Whenever the next step is an asynchronous wait — CI
running, a human decision pending — /drive reports the state and returns.
The repeat-until-done behavior is owned by the native /goal driver. You run:
/goal "<ticket-id> is merged"
and the host re-invokes /drive <ticket-id> each turn until the ticket is merged
and Done. Keeping the loop in native /goal//loop (not here) is what lets every
round run as a fresh, fully-budgeted interactive turn — this is the autopilot ADR
decision (spec/decisions / the native-/goal ADR that reframed this skill from
/goal to /drive). /drive is the recipe; /goal is the engine.
Re-entrancy — resume, never redo
Because /goal re-invokes /drive many times for the same ticket, /drive must
be idempotent per stage: on each entry it reconstructs where the ticket
already is and resumes from the furthest completed stage. It never re-claims an
already-claimed ticket and never redoes a committed stage.
Stage signals — durable first, note second
/drive infers a coarse current stage from DURABLE signals (the source of
truth), then reconciles its scratch note against them:
- unclaimed — tracker status is not In Progress and no worktree exists →
start at step 1 (
/pickup-task). - planned — an approved/committed plan file exists for the ticket → skip planning; go to implement.
- implemented — implementation commits exist on the branch beyond the plan → go to verify + review.
- in review — a PR exists (
gh pr view) → go to the merge-policy step. - parked — the ticket is
blocked_on_question(a human decision is pending) → do NOT resume the review/fix loop; report that it is parked for a human and return. A parked ticket is resumed only after its question is answered.
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.
- 7d ago First seen · 324 lines · 56 tokens per session scan A 533207f19cd4
drive is a skill published in the GitHub repository firatcand/forge (13 stars, last pushed 1mo ago), licensed MIT. It adds 56 tokens to every session and 3,811 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 2026-08-30.
Other skills, from other repositories
prospec-archive
An archiving tool for completed software changes. It creates a summary, synchronizes requirements with feature specifications, and checks that related project knowledge has been updated.
prospec-implement
An implementation workflow that reads an ordered task list and carries out the tasks one by one. Each finished task is immediately marked in tasks.md, the file holding the work list.
prospec-tasks
A task-planning workflow that turns an implementation plan and change specification into an actionable checklist.
github-triage
Triage GitHub issues through a configurable label-based state machine. Use when user wants to triage incoming issues, prepare issues for an autonomous agent, or move an issue between workflow states. Repo inferred from git remote; all GitHub calls go through gh.
to-issues
Decompose a plan, PRD, or spec into independently-grabbable vertical-slice issues (markdown file by default, GitHub issues via flag). Trigger when the user wants implementation tickets, work decomposition, or to convert an implementation plan into parallelizable work. Takes a plan file and emits atomic vertical slices.
loop-build
(loop-engineering) Implements exactly one feature from the BACKLOG.md Backlog — syncs main, acquires the loop lock, creates a feature branch, implements all acceptance criteria with tests, visually inspects UI changes, passes the independent verifier, and opens a PR. Use for Build-mode loop runs, or when asked to…