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/ProjectDXAI/labratWrote 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/projectdxai/labrat/synthesize)<a href="https://agentmods.dev/commands/projectdxai/labrat/synthesize"><img src="https://agentmods.dev/badge/commands/projectdxai/labrat/synthesize/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/projectdxai/labrat/synthesize"><img src="https://agentmods.dev/badge/commands/projectdxai/labrat/synthesize.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 | $0.00000 | $0.00190 |
| Opus 5.5 | $0.00000 | $0.00076 |
| Sonnet 5.5 | $0.00000 | $0.00038 |
| Haiku 4.5 | $0.00000 | $0.00019 |
Grade A, and why
synthesize 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 2mo 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
Synthesize what the recent evaluations mean before dispatching more work. This step closes the gap between "config sweeper" and "research agent."
- Read the last ten entries in
state/evaluations.jsonland their corresponding records instate/candidates.jsonl. - Note the distribution of
failure_classvalues and the Pareto context ifstate/pareto.jsonexists. - Answer three questions in one compact paragraph each:
- What did we learn from the last ten evaluations that is not already in
coordination/prioritized_tasks.md? - Does this change which family should earn the next credit mint?
- Which decisive challenge is the next credible unlock, and which family is closest to winning it?
- What did we learn from the last ten evaluations that is not already in
- Write your synthesis to
coordination/prioritized_tasks.md, overwriting the prior content.
Do not dispatch new work in this step. The supervisor does that next.
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.
- 2mo ago First seen · 12 lines · 0 tokens per session scan A bed93f11d67e
synthesize is a command published in the GitHub repository ProjectDXAI/labrat (236 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 190 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 commands, from other repositories
infogenius
Research-First Image Generation - Facts → Visual Prompt → High-Quality Image.
music-suno-prompt
Grounded Suno prompt synthesis from local knowledge corpus + persona canon + label canon. No vibes-prompting.
close-track
Close a completed track — run quality gate, update conductor state, commit, and optionally handle the git branch.
audit-plugin
Audit plugin skills, commands, and agents for structure, size, and naming issues.
release
Prepare a release (step 1 of 2) — version bump and PR. Cutting the tag is a separate operator-only step.
resume
Auto-detect and resume any interrupted Plan Cascade task. Detects mega-plan, hybrid-worktree, or hybrid-auto context and routes to the appropriate resume command.