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/dsswift/ion/squashgit clone --depth 1 https://github.com/dsswift/ionWhat 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 | $0.00017 | $0.04362 |
| Opus 5 | $0.00009 | $0.02181 |
| Sonnet 5 | $0.00003 | $0.00872 |
| Haiku 4.5 | $0.00002 | $0.00436 |
Grade A, and why
squash 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 yesterday.
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 — 455 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/squash
Rebuild current feature branch into clean conventional commits without changing tracked content. This command is history-only and fully autonomous.
No human interaction
Never call AskUserQuestion. Never pause for approval, confirmation, or
disambiguation at any point. Emit the plan, then execute it in the same turn
without waiting. Every decision has a deterministic rule below; when a rule
fires, apply it, log it as a plan revision, and continue.
The only stop conditions are the Phase A guards (§2), a pre-reset target mismatch (§6), and verified content loss (§8). Once the rebuild starts it always runs to completion (§7b) — a plan that proves unimplementable is revised, never abandoned. Any stop is automatic and reported; it never asks.
Absolute invariants
Two, both mandatory. Neither is ever traded for the other.
I1 — Content. Final
HEAD^{tree}must equal startingHEAD^{tree}byte-for-byte.I2 — Scope. Every result commit must resolve to exactly one scope under
.commit.json, and itstype(scope):tag must be that scope.
I2 is not cosmetic. Release-please versions each component by the paths a
commit touches (release-please-config.json), and takes the commit's type and
subject as that component's changelog entry. So a commit spanning two release
units bumps both and files the same subject under both changelogs — one of
them a lie. A commit whose tag disagrees with its paths ships a correct version
under a misleading entry. Both are defects equal in severity to losing content.
.commit.json keeps the tag and the paths in agreement, which is why scope is
always resolved from paths and never taken from the plan's subject line. There
is no "report the mismatch and move on" path: a scope violation is prevented
before commit (§7c) and repaired after commit (§8a).
Commit count, subjects, and grouping are all negotiable to satisfy I1 and I2. Splitting a feature into more commits, or merging features into fewer, is always permitted and preferred over shipping a wrong scope.
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.
- yesterday First seen · 455 lines · 17 tokens per session scan A 3b2bd65cbcb7
squash is a command published in the GitHub repository dsswift/ion (4 stars, last pushed yesterday), licensed MIT. It adds 17 tokens to every session and 4,362 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
memory-self-review
Mine recent agent history (claude-mem + usage stats) for recurring failures and repeated patterns, audit MEMORY.md health, and PROPOSE (never apply) CLAUDE.md/memory edits for human approval.
code-review
Provide a code review for the given pull request.
echo
This is a placeholder command definition for plugin framework tests.
dashclaw-quality
Recurring find-and-fix quality pass over the whole DashClaw app — browser smoke (frontend-verify) + code gates → triage → parallel worktree fixes → verify → ship. The goal prompt that drives the dashclaw-find-and-fix workflow.
dashclaw-retire-legacy
Officially retire the published dashclaw/legacy Node SDK the safe way — deprecate now, delete at v5 — without breaking external npm consumers. Drives the legacy-sdk-deprecation-sweep workflow.
apply-self-review
Apply the approved improvements from the latest memory-self-review PROPOSAL. Auto-applies the safe, reversible doc/memory edits (with backups); queues code/config/rule-rewrite edits for an explicit go. Never commits or pushes.