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.
git clone --depth 1 https://github.com/iliaal/whetstoneWrote 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/ia-compound-refresh)<a href="https://agentmods.dev/commands/iliaal/whetstone/ia-compound-refresh"><img src="https://agentmods.dev/badge/commands/iliaal/whetstone/ia-compound-refresh/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/ia-compound-refresh"><img src="https://agentmods.dev/badge/commands/iliaal/whetstone/ia-compound-refresh.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 | $0.00023 | $0.01809 |
| Opus 5 | $0.00012 | $0.00905 |
| Sonnet 5 | $0.00005 | $0.00362 |
| Haiku 4.5 | $0.00002 | $0.00181 |
Grade A, and why
ia-compound-refresh 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compound Refresh
Review institutional knowledge in docs/solutions/ for drift and staleness.
Process
1. Inventory
List all files in docs/solutions/ recursively. For each file, read the YAML frontmatter and first 20 lines of content to understand what it documents.
If docs/solutions/ doesn't exist or is empty, report that and stop.
2. Validate references
For each learning doc, check whether the code it references still exists:
- File paths mentioned in the doc -- do they still exist?
- Function/class names -- grep for them in the codebase
- Patterns described -- are they still the current approach?
- Dependencies/versions -- still accurate?
Unverifiable is not false. These are existence checks, and a repo rarely witnesses its own operations. A learning about a database tuning practice, a deploy runbook, an environment quirk, or an onboarding step has no greppable in-repo referent and never will. Act on contradiction -- the repo shows something different from what the doc claims -- not on absence of corroboration, and note the verification gap in the report instead of resolving it against the doc.
Missing files prove the implementation is gone, not the problem. If the application still deals with what the doc addresses, that is Replace, not Archive. A doc that never referenced in-repo code cannot satisfy "implementation gone" and must never auto-archive on that basis.
Mechanics follow the code; evidenced guidance does not. A doc claim about how the system currently works (a path, a function name, a config value) follows the code -- when they disagree, the doc is stale. A practice the doc justifies on its own evidence (a measured fix, a post-incident rule, a vendor constraint) does not become false because the implementation stopped satisfying it. Classify that doc from its own evidence -- usually Keep -- and report the code's drift as a potential regression in the Step 4 report instead of rewriting the practice to match broken code. Refresh edits docs only; it never adjudicates or edits product code.
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 · 120 lines · 23 tokens per session scan A 3f6ef5d6ed98
ia-compound-refresh is a command published in the GitHub repository iliaal/whetstone (33 stars, last pushed 2d ago), licensed MIT. It adds 23 tokens to every session and 1,809 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 2026-09-07.
Other commands, from other repositories
self-improve
Bootstrap and operate a self-improving agent workspace. Scaffolds .learnings/ and memory/ directories, captures errors and learnings during a session, detects recurring patterns, and promotes stable entries to project memory (CLAUDE.md, AGENTS.md, or references/). Also implements the Proactive Agent pillars — WAL…
promote
Move .memory/ artifacts to permanent knowledge locations.
checkpoint
Create or update .memory/ with current conversation state.
cleanup
Delete promoted content from the .memory/ directory.
memory-graph
Open the memory graph visualization in your browser.
f5-ctx
Context and session management - save, restore, optimize.