Borrowing it
Nothing to install: this file belongs to sorryorc/mcp-codex-worker. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sorryorc/mcp-codex-worker/main/.claude/skills/worker-research-frontier/SKILL.mdgit clone --depth 1 https://github.com/sorryorc/mcp-codex-workerWrote 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/sorryorc/mcp-codex-worker/worker-research-frontier)<a href="https://agentmods.dev/skills/sorryorc/mcp-codex-worker/worker-research-frontier"><img src="https://agentmods.dev/badge/skills/sorryorc/mcp-codex-worker/worker-research-frontier.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 5.1 | $0.00034 | $0.00365 |
| Opus 5 | $0.00017 | $0.00182 |
| Sonnet 5 | $0.00007 | $0.00073 |
| Haiku 4.5 | $0.00003 | $0.00036 |
Grade A, and why
worker-research-frontier 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 7d 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
Worker Research Frontier
Candidate Frontiers
- Dynamic tool router: expose fewer tools to cheap models.
- Compound tools: wrap search-read-edit-test into audited primitives.
- Trajectory eval: score the path, not just the final result.
- Semantic gate: detect "checks passed but meaning is wrong".
- Skill distillation: strong sessions create verified runbooks for cheap sessions.
- Context decay meter: track failure rate against prompt and output size.
External Signals
- Zhihu, "Claude Fable 5 系统泄漏？提示词全拆解" (
https://zhuanlan.zhihu.com/p/2048197182205998344): use only as a secondary structural signal. The public summary says the alleged prompt is large, heavily sectioned, and policy/tool oriented; the actionable lesson is to distill durable project doctrine into skills, validators, and gates instead of stuffing one huge prompt into every worker run. - Do not copy alleged leaked system prompt text into this repository. Extract design patterns only after restating them as repo-specific, testable rules.
Falsifiable Result
A frontier idea has a result only when it improves a measured eval set without increasing blocking false positives beyond the chosen threshold.
When Not To Use
Do not use this for immediate production fixes. Use worker-campaign-cheap-reliability.
Verification
- Add an eval fixture or metric before changing defaults.
- Run
npm testand the relevant stats/audit command.
Provenance and maintenance
Label unproven ideas as candidate until measured on this repository.
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.
- 7d ago First seen · 38 lines · 34 tokens per session scan A d1de38c9e6a7
worker-research-frontier is a skill published in the GitHub repository sorryorc/mcp-codex-worker (2 stars, last pushed 5d ago), licensed MIT. It adds 34 tokens to every session and 365 once invoked, about $0.0002 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
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
memstack-business-gdpr
Use this skill when the user says 'GDPR', 'data protection', 'privacy compliance', 'DPA', 'DSAR', 'data subject request', 'cookie consent', 'privacy audit', 'CCPA', or asks 'do I need GDPR for this repo'. Scans the repository to detect what personal data is collected, classifies sensitivity, determines whether GDPR…
memstack-development-refactor-planner
Use this skill when the user says 'refactor', 'refactoring plan', 'code cleanup', 'reduce duplication', 'simplify code', 'tech debt', 'god class', 'tight coupling', or needs to systematically improve existing code. Identifies targets, assesses risk, and builds incremental execution plans. Do NOT use for writing new…
memstack-automation-webhook-designer
Use this skill when the user says 'webhook', 'webhook handler', 'webhook endpoint', 'receive events', 'HMAC verification', 'idempotency', or needs secure webhook handlers with signature verification, retry handling, and dead letter queues. Do NOT use for full n8n workflows or scheduled tasks.
memstack-deployment-domain-ssl
Use this skill when the user says 'setup domain', 'configure DNS', 'SSL certificate', 'domain-ssl', 'custom domain', 'HTTPS setup', or needs to configure DNS records, SSL certificates, and custom domains for any hosting provider. Do NOT use for full deployment workflows.
memstack-product-feedback-analyzer
Use this skill when the user says 'analyze feedback', 'feedback analysis', 'what are customers asking for', or has support tickets, reviews, or survey data to categorize, score, and prioritize into actionable reports. Do NOT use for competitor analysis or market research.