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 adtn0810/truestack --skill truestack-prompt-optimizergit clone --depth 1 https://github.com/adtn0810/truestackWrote 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/adtn0810/truestack/truestack-prompt-optimizer)<a href="https://agentmods.dev/skills/adtn0810/truestack/truestack-prompt-optimizer"><img src="https://agentmods.dev/badge/skills/adtn0810/truestack/truestack-prompt-optimizer/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/adtn0810/truestack/truestack-prompt-optimizer"><img src="https://agentmods.dev/badge/skills/adtn0810/truestack/truestack-prompt-optimizer.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 | $0.00155 | $0.02196 |
| Opus 5 | $0.00077 | $0.01098 |
| Sonnet 5 | $0.00031 | $0.00439 |
| Haiku 4.5 | $0.00015 | $0.00220 |
Grade A, and why
truestack-prompt-optimizer 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 11d 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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
truestack-prompt-optimizer
A weak prompt produces weak work. Left raw, "fix the login thing" hands the debugger a circular echo — goal: fix the login thing; done when: the login thing is fixed — with no persona, no technique, and every gap filled silently downstream. This skill is the interception pass between a raw ask and the work: classify the task, judge what the prompt is missing, ask or infer, pick the technique, rewrite. Same intent, optimized prompt.
When to run (and when not)
- Automatically at every handoff — when truestack-orchestrate dispatches a request to a downstream skill, optimize the handoff prompt first; in a chain, re-optimize at each step for that step's destination.
- On request — "sharpen this prompt", "optimize my prompt", "adopt an expert persona".
- Degrade gracefully — if the prompt is already precise and complete, say so in one line and make only minimal edits; optimizing ceremony onto a sharp one-liner is its own failure. For a pass-through, the brief is the persona line + the prompt + the verbatim tag — the full §5 template applies only when something needed fixing.
1. Classify the task type
One of: reasoning/analysis · factual lookup · classification · extraction · generation/creative · coding · transformation. State it with a one-line reason. Classification here picks the technique, never the destination — the router already matched the request to its skill; don't re-derive the route, don't second-guess it, and don't do the work here. truestack-orchestrate owns dispatch.
2. Judge the four components — needed vs present
Not every task needs all four; judge which are required for this one, then mark each present · inferable · missing-and-required.
| Component | What it is | Usually required for |
|---|---|---|
| Instruction | The explicit action to perform | Every task |
| Context | Background, constraints, audience, tone, goal | Generation, analysis, anything audience-facing |
| Input data | The material to operate on | Extraction, transformation, classification, debugging |
| Output shape | Format, length, structure, or schema | Structured output, extraction, transformation |
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.
- 11d ago First seen · 147 lines · 155 tokens per session scan A 39ae78831677
truestack-prompt-optimizer is a skill published in the GitHub repository adtn0810/truestack (2 stars, last pushed 2mo ago), licensed MIT. It adds 155 tokens to every session and 2,196 once invoked, about $0.0008 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
guidance
Constrain LLM output with grammars; guarantee valid JSON.
outlines
Outlines: structured JSON/regex/Pydantic LLM generation.
instructor
Structured LLM outputs validated with Pydantic.
dspy
DSPy: declarative LM programs, auto-optimize prompts, RAG.
claude-api
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude…
omni-compression
Configure RTK (command output), Caveman (prose), and stacked compression modes. Manage language packs, custom rules, and test prompt compression reducing tokens by 60–90%.