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/iliaal/whetstonenpx agentmods add commands/iliaal/whetstone/announceWrote 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/iliaal/whetstone/announce)<a href="https://agentmods.dev/commands/iliaal/whetstone/announce"><img src="https://agentmods.dev/badge/commands/iliaal/whetstone/announce/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/commands/iliaal/whetstone/announce"><img src="https://agentmods.dev/badge/commands/iliaal/whetstone/announce.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 | $0.00017 | $0.02452 |
| Opus 5.5 | $0.00007 | $0.00981 |
| Sonnet 5.5 | $0.00003 | $0.00490 |
| Haiku 4.5 | $0.00002 | $0.00245 |
Grade A, and why
announce 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 29d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Draft release announcement for X
Generate an announcement post for the latest whetstone plugin release. The account is X Premium — one long post usually covers a release; only thread when a second distinct theme justifies the visual pacing.
This command is the dedicated X drafter for whetstone releases. For other repos, other platforms, or non-release posts (benchmark drops, opinion takes, real-time engagement), use /ia-promote <repo> <platform> instead.
Step 0: Load promotion-workspace rules
Before drafting, read these to anchor voice, platform constraints, and ethics:
~/ai/promotion/rules/platforms/x.md— X-specific rules (Premium char limits, hook discipline, when to thread vs single-post)~/ai/promotion/rules/voice.md— voice rules (directness, density, banned constructions)~/ai/promotion/rules/ethics.md— non-negotiables (no shadow personas, no manufactured urgency, no fake testimonials, no manufactured social proof)~/ai/promotion/rules/humanizer-check.md— anti-AI-detection patterns specific to social posts
When this command's inline guidance disagrees with those rules, the rules win — the rules are the canonical source, this command is the orchestration wrapper. Apply the ia-writing skill throughout for prose hygiene (no filler, no AI slop, no throat-clearing).
Also check ~/ai/promotion/announcement-matrix.md for the whetstone row. If Last announced already equals the version being drafted, ask the user whether to skip, redraft anyway, or target a different version.
Step 1: Gather context
- Read the current version from
plugins/whetstone/.claude-plugin/plugin.json - Read CHANGELOG.md -- combine ALL entries newer than the matrix's
Last announcedversion for X (from~/ai/promotion/announcement-matrix.md, read in Step 0) into one announcement. The announcement covers the full gap, not just the newest version: the matrix can lag several versions behind current, and every unannounced version's user-visible changes belong in this post. IfLast announcedalready equals the current version, there is nothing new to announce (see Step 0). - Get the ai-skills repo version from its sync commit:
cd ~/ai/ai-skills && git log --oneline --grep '^sync: v' -1 && cd - - Count components (read-only, no writes):
bash scripts/update-metadata.sh --dry-run
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.
- 29d ago Changed 86052b35bfbc
- 1mo ago Changed 9a375685231b
- 1mo ago First seen · 135 lines · 17 tokens per session scan A ab66f9a2456a
announce is a command published in the GitHub repository iliaal/whetstone (35 stars, last pushed yesterday), licensed MIT. It adds 17 tokens to every session and 2,452 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
sm-buff
You usually don't run buff directly — run sm sail. sm sail drives the whole PR to green and calls buff watch / triage for you, stopping only when it needs you to act (see /sm-sail). Reach for sm buff here only for surgical work: inspecting a specific failure, or resolving a single review thread when sail has parked on…
sm-scour
Run slop-mop's comprehensive pre-PR sweep for this repository.
repo-audit
Read-only quality audit summary via repoindex ops audit. Shows failing checks by severity without fixing anything.
review
Multi-agent code review with parallel validation.
speckit.boundedrelay.plan-review
Freeze Claude host findings before an independent BoundedRelay Codex plan review.
speckit.boundedrelay.implementation-review
Run frozen Claude host and independent Codex implementation reviews.