Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/V-Songbird/slagnpx agentmods add skills/v-songbird/slag/check-writerWrote 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/v-songbird/slag/check-writer)<a href="https://agentmods.dev/skills/v-songbird/slag/check-writer"><img src="https://agentmods.dev/badge/skills/v-songbird/slag/check-writer/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/v-songbird/slag/check-writer"><img src="https://agentmods.dev/badge/skills/v-songbird/slag/check-writer.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 | $0.00105 | $0.01600 |
| Opus 5.5 | $0.00042 | $0.00640 |
| Sonnet 5.5 | $0.00021 | $0.00320 |
| Haiku 4.5 | $0.00011 | $0.00160 |
Grade A, and why
check-writer 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 22d 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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Check writer
A check is admitted on one basis and no other: it fires on its own violation fixture and stays silent on its own near miss. Nothing else counts — not who wrote it, not how sensible it sounds, not how many other checks passed beside it. A second pattern is not covered because the pattern next to it matched.
On Antigravity, give every command the project root the person named as its working directory, the
Cwd of run_command. Without it, commands can run in Antigravity's own scratch directory instead
of the project. When the person named no project root, ask for it before running anything.
First, look for an open task
Run node .collet/task.mjs status before writing anything. While a task is open, the session guard
refuses writes under .collet/ other than .collet/unverified.md: the checks, their fixtures and
the ledger are the harness's own state, not task files. Widening cannot lift that refusal, and
task.mjs widen refuses those paths.
If a task is open, tell the person before any write is refused, and add the check between tasks:
- Finish the open task and close it:
node .collet/task.mjs close --left-out "..." --unverified "...". A task whose accept command does not pass yet cannot close; name the missing check in--left-outwhen it does close, or in your summary, so it is not lost. - With no task open, follow the steps below, admission included.
- Open the next task. The check runs from then on, when a file is written and when a task closes.
Do not work around the refusal. The guard also refuses a shell command that would create a new file
under .collet/, touch, ni and mkdir included, a .collet path in another letter case, and
removing or moving .collet itself. It reads only the shell forms it recognizes, though, and the
close-time scope check skips that directory. A check that gets there another way is never
reviewed, yet it runs when the open task closes, and one without a working live keeps that task
open. .collet/off silences every session guard, the open task's scope included. Switching it on
is the person's decision, not a step of this skill.
What ships with it
1 file 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.
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.
- 22d ago Changed · +28 lines 9bf19ae36830
- 23d ago First seen · 104 lines · 105 tokens per session scan A df48358fa5d5
check-writer is a skill published in the GitHub repository V-Songbird/slag (1 stars, last pushed 6d ago), licensed MIT. It adds 105 tokens to every session and 1,600 once invoked, about $0.0004 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 2026-09-23.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.