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 commands/rolecraft-sh/rolecraft/watchgit clone --depth 1 https://github.com/rolecraft-sh/rolecraftWrote 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/commands/rolecraft-sh/rolecraft/watch)<a href="https://agentmods.dev/commands/rolecraft-sh/rolecraft/watch"><img src="https://agentmods.dev/badge/commands/rolecraft-sh/rolecraft/watch.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 | $0.00000 | $0.00348 |
| Opus 5 | $0.00000 | $0.00174 |
| Sonnet 5 | $0.00000 | $0.00070 |
| Haiku 4.5 | $0.00000 | $0.00035 |
Grade A, and why
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 4d 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.
What it actually says
rolecraft watch
Watch installed skills for file changes and auto-sync to agent directories.
Usage
rolecraft watch [<slug>] [--dry-run]
Options
| Flag | Description |
|---|---|
--dry-run |
Preview which skills would be watched without starting file watchers |
Description
Monitors skill source directories using node:fs FSWatcher. When a file changes, the skill is automatically re-installed to all its configured agent targets and the lockfile is updated with new content hashes.
Only local source skills can be watched. Remote (GitHub, npm, git) sources are skipped.
Examples
# Watch all local skills
rolecraft watch
# Watch a specific skill
rolecraft watch my-skill
# Preview without starting watchers
rolecraft watch --dry-run
Output
👀 Watching 2 skill(s) for changes...
✓ my-skill → watching /Users/me/projects/my-skill
✓ team-rules → watching /Users/me/projects/team-rules
📌 Press Ctrl+C to stop watching
On file change:
[2:30:15 PM] my-skill: SKILL.md changed, syncing...
[2:30:15 PM] my-skill: synced successfully
Node.js API
This command is also available as a programmatic function. See the Node.js API documentation for detailed usage.
import { watch } from 'rolecraft'
const { watchers, skills } = await watch('my-skill', process.cwd())
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.
- 4d ago First seen · 62 lines · 0 tokens per session scan A 4ac7cc99be06
watch is a command published in the GitHub repository rolecraft-sh/rolecraft (75 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 348 tokens. 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 commands, from other repositories
audit-artifacts
List suspicious root agent process files and placement directories.
audit
Diff live infrastructure against the spec and report drift by component name.
uninstall
Uninstall filetree from this repo — un-wires CLAUDE.md/AGENTS.md, removes generated artifacts (confirmed per item), tears down self-only mode if enabled.
research-component
Research and add a new infra component from a documentation link the user provides — crawl the docs, fan out subagents over the relevant subpages, then write, register, and validate the component.
vp-get-active-cell
VP Get active cell ( vpAreaName : Text { ; sheet : Integer } ) : Object.
vp-get-frozen-panes
VP Get frozen panes ( vpAreaName : Text { ; sheet : Integer } ) : Object.