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 2389-research/simmer --skill simmer-reflectgit clone --depth 1 https://github.com/2389-research/simmerWrote 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/2389-research/simmer/simmer-reflect)<a href="https://agentmods.dev/skills/2389-research/simmer/simmer-reflect"><img src="https://agentmods.dev/badge/skills/2389-research/simmer/simmer-reflect/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/2389-research/simmer/simmer-reflect"><img src="https://agentmods.dev/badge/skills/2389-research/simmer/simmer-reflect.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 | $0.00059 | $0.02013 |
| Opus 5 | $0.00030 | $0.01007 |
| Sonnet 5 | $0.00012 | $0.00403 |
| Haiku 4.5 | $0.00006 | $0.00201 |
Grade A, and why
simmer-reflect 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.
How it starts
The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Simmer Reflect
You are the only subskill that sees the full score history. Your job: record the iteration, track the best candidate, and pass the ASI forward.
Context You Receive
- Full score history: all iterations so far (scores, composites, key changes)
- Current iteration number and max iterations
- Latest judge output: scores + ASI for this round
- Generator report: what changed this round (2-3 sentences)
- Artifact type: single-file or workspace
What To Do
1. Record in Trajectory
Update {OUTPUT_DIR}/trajectory.md with the running score table.
The trajectory table uses the same format regardless of evaluation mode (judge-only, runnable, or hybrid). Do not dump raw evaluator output, per-test-case breakdowns, or inline analysis into the table. The table is a clean score record.
Required format (do not add extra columns beyond those listed below):
Single-file mode:
- Columns: Iteration, [criterion names from rubric], Composite, Key Change
- Below table:
Best candidate: iteration [N] (composite: [N.N]/10)
# Simmer Trajectory
| Iteration | [criterion 1] | [criterion 2] | [criterion 3] | Composite | Key Change |
|-----------|---------------|---------------|---------------|-----------|------------|
| 0 | 4 | 5 | 3 | 4.0 | seed |
| 1 | 7 | 5 | 4 | 5.3 | [summary] |
| 2 | 7 | 6 | 7 | 6.7 | [summary] |
Best candidate: iteration 2 (composite: 6.7/10)
Workspace mode:
- Columns: Iteration, [criterion names from rubric], Composite, Config, Key Change
- The "Config" column captures the key workspace state (model name, topology, etc.) in a few words
- Below table:
Best candidate: iteration [N] (composite: [N.N]/10)
# Simmer Trajectory
| Iteration | Coverage | Efficiency | Noise | Composite | Config | Key Change |
|-----------|----------|------------|-------|-----------|--------|------------|
| 0 | 2 | 6 | 2 | 3.3 | qwen3.5:4b, single-call | seed |
| 1 | 4 | 4 | 4 | 4.0 | qwen3.5:27b, single-call | model swap + rich prompt |
| 2 | 3 | 5 | 3 | 3.3 | qwen3.5:9b, single-call | REGRESSION — 9b too weak |
Best candidate: iteration 1 (composite: 4.0/10)
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 · 178 lines · 59 tokens per session scan A 7bec87dffce1
simmer-reflect is a skill published in the GitHub repository 2389-research/simmer (14 stars, last pushed 2mo ago), licensed MIT. It adds 59 tokens to every session and 2,013 once invoked, about $0.0003 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 skills, from other repositories
sillytavern-iterator
A writing workflow that takes work through drafting, review, revision, polishing, and finalising. It also uses checks such as reading error signals, searching for answers, and testing assumptions.
kaizen
Kaizen (改善): every process has waste; eliminate it incrementally.
dag-feedback-synthesizer
Synthesizes actionable feedback from validation results, confidence scores, and iteration triggers. Creates structured improvement guidance for re-execution. Activate on 'synthesize feedback', 'improvement suggestions', 'actionable feedback', 'iteration guidance', 'feedback generation'. NOT for iteration detection…
datamodel-code-generator
Use this skill when the user wants Python data models, Pydantic models, dataclasses, TypedDicts, msgspec structs, or type-safe Python classes generated from OpenAPI, AsyncAPI, JSON Schema, GraphQL, JSON/YAML/CSV sample data, MCP tool schemas, Protocol Buffers, XML Schema, Apache Avro, or existing Python model objects.…
trulens-diagnosis
Diagnose low evaluation scores and generate actionable improvement recommendations.
state-trees
Create, inspect, and edit StateTree assets — states, tasks, transitions, conditions, considerations, and property bindings for AI behavior and game logic (StateTreeService). Use when the user asks to build a StateTree, add states/tasks/transitions, bind task properties, set up Utility AI considerations, or wire…