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/vitaliikapliuk/modelharnessWrote 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/vitaliikapliuk/modelharness/retro)<a href="https://agentmods.dev/commands/vitaliikapliuk/modelharness/retro"><img src="https://agentmods.dev/badge/commands/vitaliikapliuk/modelharness/retro/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/vitaliikapliuk/modelharness/retro"><img src="https://agentmods.dev/badge/commands/vitaliikapliuk/modelharness/retro.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 | $0.00018 | $0.00146 |
| Opus 5 | $0.00009 | $0.00073 |
| Sonnet 5 | $0.00004 | $0.00029 |
| Haiku 4.5 | $0.00002 | $0.00015 |
Grade A, and why
retro 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 9d 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
Run a short retrospective of this session and persist what is worth remembering.
- Load the
memory-disciplineskill and follow its write rules. - Identify candidate lessons: user corrections, approaches that worked after failed attempts, non-obvious project facts discovered the hard way. Exclude anything the repo already records.
- For each lesson: create/update a file in
lessons/(one lesson per file, one-line summary first), then refreshMEMORY.mdindex lines. - Show the user a one-line-per-lesson summary of what was saved.
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.
- 9d ago First seen · 14 lines · 18 tokens per session scan A ad7f95221ab9
retro is a command published in the GitHub repository vitaliikapliuk/modelharness (38 stars, last pushed 2mo ago), licensed MIT. It adds 18 tokens to every session and 146 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 2026-08-30.
Other commands, from other repositories
recall-save
Generate / overwrite .recall/context.md with Recall's local offline summarizer.
learn
Force claude-smart to extract learnings from this session now.
ccr
CCR (Compressed Context with Retrieval) — recall the full original of context that greatcto compressed/filtered out, by its short id. The retrieval half of the compression layer.
ai-act-ask
Answer an EU AI Act question grounded in the bundled knowledge base — verbatim statute text, obligation paraphrases, and the compound-risk taxonomy. Offline and deterministic by default; cites the articles it relies on.
broadcast
Run broadcast on a distilled Raw — update related existing pages conversationally.
save
Save the current session digest. Fans out across every writable memory tier per the house-map. Tier 0 always; Tier 1 if reachable; Tier 2 only if registered writable with auth. Each tier independent — failures degrade gracefully.