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 skills add hyhmrright/logic-lens --skill iterate-skillgit clone --depth 1 https://github.com/hyhmrright/logic-lensWrote 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/hyhmrright/logic-lens/iterate-skill)<a href="https://agentmods.dev/skills/hyhmrright/logic-lens/iterate-skill"><img src="https://agentmods.dev/badge/skills/hyhmrright/logic-lens/iterate-skill/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/hyhmrright/logic-lens/iterate-skill"><img src="https://agentmods.dev/badge/skills/hyhmrright/logic-lens/iterate-skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00000 | $0.01766 |
| Opus 5.5 | $0.00000 | $0.00706 |
| Sonnet 5.5 | $0.00000 | $0.00353 |
| Haiku 4.5 | $0.00000 | $0.00177 |
Grade A, and why
iterate-skill 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 1mo 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
iterate-skill — the Logic-Lens improvement loop
This is the harness orchestrator. It coordinates three agents and two support skills into one deterministic loop that raises a skill's eval score without overfitting or grader-gaming.
Execution mode: sub-agent pipeline (generate → test → verify). Each step's output is the next
step's input, handed off through the filesystem (skills-workspace/iteration-<TAG>/) and agent
return values. There is no peer-to-peer team chatter, so agents are spawned via the Agent tool —
always with model: "opus", and subagent_type set to the agent's own definition name (e.g.
subagent_type: "skill-editor"), not "general-purpose" — passing general-purpose would discard
the role/principles in .claude/agents/<name>.md, defeating the harness. Agents return results to
this orchestrator, which owns state and the ship/rollback decisions.
Agents (who) — all in .claude/agents/, spawn by these exact subagent_type names:
subagent_type |
Role |
|---|---|
eval-failure-analyzer |
Read-only: cluster failures, map to eval IDs, propose minimal edits |
skill-editor |
Apply one minimal, generalized edit; refuse grader edits |
iteration-guard |
Verify net gain vs variance; recommend SHIP / ROLLBACK / RERUN (orchestrator executes any revert) |
Support skills (how): sync-skill-cache (mandatory pre-eval gate), run-iteration-eval (run + grade).
Phase 0 — context check (initial / resume / partial)
Determine the run mode before doing anything:
ls -dt skills-workspace/iteration-*/— if recent iterations exist and the user asks to "continue" or "another pass" → resume: use the latest as baseline, skip re-baselining.- User provides a fresh target skill / new failure → initial: establish a baseline first (Phase 1).
- User asks to redo just one mode or one skill → partial: scope the eval to the affected case IDs.
Confirm the target skill (which of the six logic-*) and the failing mode with the user if ambiguous —
do not guess which skill to mutate.
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.
- 1mo ago First seen · 118 lines · 0 tokens per session scan A 22cc9b50c1bb
iterate-skill is a skill published in the GitHub repository hyhmrright/logic-lens (25 stars, last pushed 25d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,766 tokens. 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 skills, from other repositories
brooks-health
Combined codebase health dashboard that scores a project across all four quality dimensions — PR quality, architecture, tech debt, and test quality — in a single pass, drawing on twelve classic engineering books. Triggers when: user wants an overall quality assessment, asks "how healthy is this codebase?", "run all…
brooks-sweep
Full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic engineering books. Triggers when: user…
brooks-test
Test quality review drawing on twelve classic engineering books — with primary focus on xUnit Test Patterns, The Art of Unit Testing, How Google Tests Software, and Working Effectively with Legacy Code — that diagnoses structural problems in an existing test suite: brittleness, mock abuse, coverage illusions, slow…
mutation-test
Mutation testing with two engines. Uses the project's NATIVE mutation runner (StrykerJS / Infection / mutmut / PIT / cargo-mutants) when one is configured — installing it on explicit consent when it is not — for a reproducible, comparable score; and an LLM-guided engine for the mutation classes native mutators cannot…
write-tests
Write tests for existing production code. Processes ONE file at a time through a full pipeline: analyze, inventory (frozen BEFORE writing), write, executable coverage gate, verify, blind coverage audit, adversarial review, log. Uses CodeSift for discovery and analysis when available. Modes: [path] (specific target)…
code-audit
Batch audit of production files against CQ1-CQ40 quality gates and CAP1-CAP29 anti-patterns. Tiered output (A/B/C/D), critical gate enforcement, evidence-backed scoring, cross-file pattern analysis, and prioritized execution plan. Flags: zuvo:code-audit all | [path] | [file] | --deep | --quick | --services |…