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 GregoryHo/arcforge --skill executinggit clone --depth 1 https://github.com/GregoryHo/arcforgeWrote 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/gregoryho/arcforge/executing)<a href="https://agentmods.dev/skills/gregoryho/arcforge/executing"><img src="https://agentmods.dev/badge/skills/gregoryho/arcforge/executing/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/gregoryho/arcforge/executing"><img src="https://agentmods.dev/badge/skills/gregoryho/arcforge/executing.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 | $0.00051 | $0.01687 |
| Opus 5.5 | $0.00020 | $0.00675 |
| Sonnet 5.5 | $0.00010 | $0.00337 |
| Haiku 4.5 | $0.00005 | $0.00169 |
Grade A, and why
executing 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 today.
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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Executing
Two things happen here: the work gets written down as a task list, and the list gets run. Both halves lean on the same rule — the list file is the state.
The list is the only progress record
Your context does not survive a compaction, a new session, or a handoff to a fresh subagent. A file in the repository survives all three, so completion is recorded there and nowhere else.
That means: no second ledger file, no "done so far" summary that outlives the turn, no todo tool standing in as the source of truth. A todo tool is fine as a scratchpad for the current turn; it is not the record. If two places can disagree about what is finished, one of them eventually will, and the expensive failure — redoing work that was already done — comes from believing the wrong one.
Format
# Tasks: parser rewrite
> arcforge task list v1 — status markers: `[ ]` pending, `[~]` in-progress,
> `[x]` done, `[!]` blocked. Every task has a stable id (`T1`, `T2`, …) that is
> never renumbered or reused. An indented `verify:` line is the command that
> proves the task done; a `note:` line explains a block. Edit markers in place —
> this file is the only state.
- [x] T1 — Write the failing test for the header parser
- verify: `npm run test:scripts`
- [~] T2 — Implement the parser
- [ ] T3 — Wire it into the loop
- verify: `npm test`
- [!] T4 — Publish the package
- note: registry credentials are not available in this environment
Rules that are not negotiable, because a reader with no context depends on them:
- The banner line stays. It is how a fresh worker learns the notation.
- Ids are stable. A deleted task does not renumber its neighbours; gaps are fine.
- Four markers, no fifth.
[!]requires anote:saying why — a block with no stated reason is unactionable. - Markers are edited in place. Everything else in the file — a human's comments, ordering, extra prose — is left byte for byte alone.
- No nested checkboxes. Sub-structure belongs in its own list.
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.
- today First seen · 142 lines · 51 tokens per session scan A 17fa36f1535c
executing is a skill published in the GitHub repository GregoryHo/arcforge (8 stars, last pushed 12d ago), licensed MIT. It adds 51 tokens to every session and 1,687 once invoked, about $0.0002 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 2026-11-07.
executing compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| executingGregoryHo/arcforge | Skill | 8 | 12d ago | MIT | A | 51 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |