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 3yesore/LetUenforskills --skill asa-model-comparison-judgegit clone --depth 1 https://github.com/3yesore/LetUenforskillsWrote 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/3yesore/letuenforskills/asa-model-comparison-judge)<a href="https://agentmods.dev/skills/3yesore/letuenforskills/asa-model-comparison-judge"><img src="https://agentmods.dev/badge/skills/3yesore/letuenforskills/asa-model-comparison-judge/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/3yesore/letuenforskills/asa-model-comparison-judge"><img src="https://agentmods.dev/badge/skills/3yesore/letuenforskills/asa-model-comparison-judge.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 | $0.00027 | $0.00930 |
| Opus 5 | $0.00014 | $0.00465 |
| Sonnet 5 | $0.00005 | $0.00186 |
| Haiku 4.5 | $0.00003 | $0.00093 |
Grade A, and why
asa-model-comparison-judge 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 12d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ASA Model Comparison Judge
Use this internal meta-skill when multiple LLM runs analyze the same skill and the project must compare quality, omissions, disagreements, and role suitability.
This skill fixes the failure mode where model comparison only counts output length, treats different wording as meaningful disagreement, or fails to show which model is best for each anchor type during development testing.
Inputs
- Two or more run directories for the same source skill.
- Run metadata: provider, model, source commit, config, language, and timestamps.
- Structure, workflow, reviewer, pattern, vault, and report artifacts.
- Deterministic quality metrics and reviewer issues.
Process
- Confirm compared runs use the same source repository, skill path, and preferably the same commit.
- Compare completeness across identity, trigger, resource, workflow, evidence, reuse, and bilingual layers.
- Compare evidence grounding before comparing prose quality.
- Identify real disagreements: different claims about behavior, role, boundaries, or workflow order.
- Identify omissions: useful fields present in one run but missing in another.
- Rate each model by role: structure, workflow, review, reuse, bilingual writing, and Obsidian usefulness.
- Compare anchors by type so consensus, disagreement, and best-model-per-anchor-type are visible.
- Explain whether a difference is quality, style, unsupported speculation, or harmless surface variation.
- Produce benchmark-ready tables and a cautious recommendation for development testing only.
Output Contract
Return model comparison content suitable for benchmark reports:
model_comparison:
compared_runs: []
per_model_scores: []
disagreements: []
omissions: []
best_for_role:
recommendation:
anchors:
anchor_consensus:
- id:
anchor_type:
agreed_claim:
agreeing_models: []
evidence_basis: direct | structural | inferred | mixed | unknown
confidence: high | medium | low | unknown
usable_as_baseline: true | false
notes:
anchor_disagreement:
- id:
anchor_type:
disagreement:
models_involved: []
competing_claims: []
likely_cause: missing_evidence | inference_gap | source_ambiguity | model_hallucination | style_difference | unknown
resolution_action: inspect_source | prefer_evidence_grounded | rerun_model | mark_inconclusive | ignore_style_difference
severity: high | medium | low
evidence: []
best_model_per_anchor_type:
- anchor_type:
best_model:
best_provider:
reason:
strengths: []
weaknesses: []
use_for_roles: []
avoid_for_roles: []
confidence: high | medium | low | unknown
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.
- 12d ago First seen · 112 lines · 27 tokens per session scan A c316441b5cdf
asa-model-comparison-judge is a skill published in the GitHub repository 3yesore/LetUenforskills (2 stars, last pushed 2mo ago), licensed MIT. It adds 27 tokens to every session and 930 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 skills, from other repositories
link-memory
Use after important user-approved decisions, when durable context should be proposed or reviewed, and for explicit Link memory lifecycle work: remember, recall, review, update, archive, restore, forget, or explain local memories through the CLI without requiring MCP.
link-retrieve
Use before answering work that may depend on user memory, project history, source-backed notes, or prior decisions; retrieve compact Link context through the CLI without loading the whole wiki or requiring MCP.
link-health
Use at the start of Link work when readiness is unclear, after installs or upgrades, and before repairs; verify health, inspect interrupted writes, back up, and repair generated indexes without MCP.
link-ingest
Use when raw files are present, source pages look stale, or a user asks to ingest notes into Link; refresh source-backed wiki pages, propose memories, and validate updates through the CLI without MCP.
kb
Use when compiling raw sources into the wiki, querying the knowledge base, running health checks or lint on the wiki, evolving or improving wiki coverage, or when user says 'compile', 'update the wiki', 'query', 'lint', 'health check', 'evolve', 'what's missing', or 'suggest improvements'. Also triggers on questions…
kb-init
Use when setting up a new knowledge base, bootstrapping an Obsidian vault, or when user says 'init kb', 'new knowledge base', 'create kb', or 'setup vault'. Triggers on any request to initialize or scaffold a knowledge base project.