A cross-platform workflow for academic research, from planning and reading papers to analysis, writing, proofreading, and responding to reviews. It uses task IDs and defined output files to organise work across stages.

Planning studies, reviewing literature, finding research gaps, designing methods, analysing data, writing manuscripts, and preparing rebuttals. It supports empirical, qualitative, systematic-review, methods, and theory papers.

It gives research tasks a repeatable structure and records outputs in predictable locations. It also adds review and quality checks for work intended for submission.

qiongli — Skill for Claude Code and Codex by jxpeng98

Install

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.

Any agent

npx skills add jxpeng98/qiongli --skill workflow

Clone the repo

git clone --depth 1 https://github.com/jxpeng98/qiongli

Made for: Claude Code, Codex.

Or install qiongli, the plugin that ships this one along with the rest of its 2 skills.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/jxpeng98/qiongli/workflow/github.svg)](https://agentmods.dev/skills/jxpeng98/qiongli/workflow)

Your own site

<a href="https://agentmods.dev/skills/jxpeng98/qiongli/workflow"><img src="https://agentmods.dev/badge/skills/jxpeng98/qiongli/workflow/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/jxpeng98/qiongli/workflow"><img src="https://agentmods.dev/badge/skills/jxpeng98/qiongli/workflow.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,804 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

high Tool Misuse · line 187
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
medium Tool Misuse · line 187
Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Fix: Override unsafe defaults with secure settings (verify=True, auth required, restrictive permissions). Review and harden all tool configurations.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

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	$0.00035	$0.03804
Opus 5.5	$0.00014	$0.01522
Sonnet 5.5	$0.00007	$0.00761
Haiku 4.5	$0.00003	$0.00380

Measured 28d ago against content hash 2ff7da66e6c7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

qiongli 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 28d 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.

content/workflow/SKILL.md · 261 lines

How it starts

The opening of the file, as written. The whole thing — 261 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Qiongli Academic Workflow

Help the user complete the requested academic task with traceable evidence. The active Host (Codex, Claude Code or another client) owns the model and tools; Qiongli supplies research contracts, references and project services. This is a self-contained skill package: all resource paths below are relative to this package, not the working directory.

Installed Qiongli workflow version: v2.5.3

Start with the request

First honor an active reply-only choice before reading resources or routing. The user can select no-qiongli / “NoQ问理”, or say “仅回复”, “不处理”, “no 处理”, “reply only” or “no tools”. The independent no-qiongli Skill is self-contained; do not load it or another resource merely to enter this mode. In this mode, answer from material already visible in the conversation. Do not call tools, programs, MCP, search, other agents or resource readers; do not read or write files, check connections, start project runs or refresh the Graph. Explanations, proposed text and suggested commands may be returned in chat only. If needed material is not visible, state the gap or ask the user to paste it; do not fetch it or imply that it was checked. Skip the execution defaults below, including formal-workflow gates that require tools; do not claim those gates passed.

Keep this choice for the current conversation until the user explicitly resumes execution, unless they limited it to one reply. Do not save a mode setting or call a tool to enter/leave it. A bare “no”, quoted source text, discussion/development of the entry itself, or a Hook reminder does not change the choice. An explicit request to resume actions restores ordinary routing only within its authorized scope. This is Skill guidance, not a Host-level tool lock or a way to override higher-priority instructions or stop Host hooks.

Reuse the supplied materials, conversation decisions and selected project. Infer the task internally; do not ask the user to learn a Task ID or choose a paper type already evident from context. Ask one focused question only when the missing answer changes the result, destination or permitted action.
Choose the smallest route below and read its workflow or skill card. Do not load the whole catalog, all stage playbooks or skills-core.md by default. Use skills-summary.md only when the appropriate card is unclear; the full card owns its inputs, outputs, quality bar and missing-evidence behavior.
Complete the requested unit. A question about a supplied paper or paragraph can be answered in chat. A formal task or saved artifact uses the canonical paths and gates in references/workflow-contract.md. Do not turn a quick answer into a whole project, or call it a completed formal task.
Report the result, evidence limits and any actual saved artifact. Ask for input only on the blocked portion while continuing useful authorized work.

Read the full file on GitHub · 261 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

28d ago Changed 2ff7da66e6c7
29d ago Changed 936ccc535be3
1mo ago Changed · +10 lines 6517696f14f5
1mo ago Changed 1fb49e6213fa
1mo ago Changed · +14 lines bab0ffea7217
1mo ago Changed · -47 tokens per session 91f0f2e5dfa6
1mo ago Changed · -20 lines · -22 tokens per session b355117e1191
2mo ago First seen · 257 lines · 104 tokens per session scan A 25edba0b935b

Subscribe to this mod's changes

qiongli is a skill published in the GitHub repository jxpeng98/qiongli (27 stars, last pushed 28d ago), licensed MIT. It adds 35 tokens to every session and 3,804 once invoked, about $0.0001 per session on Opus 5.5. 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.

Claude Code Research Writing and docs academic-research claude-code-plugin codex codex-plugin mcp

qiongli next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
qionglijxpeng98/qiongli	Skill	27	28d ago	MIT	A	35	Claude Code, Codex
evolveSethGammon/Citadel	Skill	927	yesterday	MIT	A	60	Claude Code
improveSethGammon/Citadel	Skill	927	yesterday	MIT	A	48	Claude Code
watchSethGammon/Citadel	Skill	927	yesterday	MIT	A	70	Claude Code
pr-watchSethGammon/Citadel	Skill	927	yesterday	MIT	A	46	Claude Code
qaSethGammon/Citadel	Skill	927	yesterday	MIT	A	56	Claude Code
reviewSethGammon/Citadel	Skill	927	yesterday	MIT	A	17	Claude Code