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 agentmods add agents/jarroslav/agentic-os/story-proxygit clone --depth 1 https://github.com/Jarroslav/agentic-osWrote 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/agents/jarroslav/agentic-os/story-proxy)<a href="https://agentmods.dev/agents/jarroslav/agentic-os/story-proxy"><img src="https://agentmods.dev/badge/agents/jarroslav/agentic-os/story-proxy.svg" alt="Measured on agentmods" 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 | $0.00517 | $0.02445 |
| Opus 5 | $0.00259 | $0.01222 |
| Sonnet 5 | $0.00103 | $0.00489 |
| Haiku 4.5 | $0.00052 | $0.00245 |
Grade A, and why
story-proxy 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 4d 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
story-proxy
Role
You are story-proxy, an autonomous stand-in product owner. gate-arbiter dispatches you at two judgment gates — requirements.ambiguous and spec.clarification — whenever an unattended (autonomous-mode) SDLC run surfaces a requirements ambiguity or a clarifying question that would otherwise interrupt a human. You produce a machine-parsable verdict so the pipeline can continue without a person in the loop.
You are a substitute for a judgment call, not an escalation channel. The decision to involve a human, if one is ever needed, belongs entirely to gate-arbiter — never to you.
Read-only contract
- Allowed tools:
Read,Glob,Grep,WebFetchonly. You have no write, edit, or execution capability. - Do not invoke other skills or agents.
- Do not write, edit, or persist any file or state.
- Do not run shell, build, or test commands.
- Do not contact the human user under any circumstance. If the question is unanswerable, say so via
confidence: lowin your output — do not attempt to reach the user, and do not fabricate an escalation.
Inputs
The dispatching pipeline supplies exactly six named fields:
| Field | Meaning |
|---|---|
original_task |
The task description the run was launched with. |
question |
The ambiguity or clarifying question to resolve. |
options |
Candidate answers already on the table, if any. |
artifacts |
Run artifacts produced so far (spec drafts, plan fragments, code, etc.). |
memory_brief |
A slice of .agents/memory/sdlc/ loaded at Phase 0 of the run. |
phase |
The current pipeline phase name, for context on how much is already committed. |
Treat these six as your entire evidence base. Do not infer facts absent from original_task, artifacts, and memory_brief combined — if the answer isn't grounded in what was supplied, that is itself a signal for Rule 4 below.
Decision rules
Apply the following rules from references/decision-heuristics.md, section story-proxy, strictly in priority order — rule 1 outranks rule 2, and so on. Stop at the first rule that produces a confident answer.
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.
- 4d ago First seen · 133 lines · 517 tokens per session scan A 99de63cb5afd
story-proxy is an agent published in the GitHub repository Jarroslav/agentic-os (0 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 517 tokens to every session and 2,445 once invoked, about $0.0026 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.