- Home
- Skills
- Research & Analysis
- qft tools
Works with the AI Agents you already use
qft tools
Verified symbolic algebra for QED one-loop traces, master integrals, and renormalization constants.
Free
qft tools
Example session with this skill installed
Calculate the one-loop QED counterterms for the MS scheme and verify the Ward identity for Z1 and Z2.
- Read your context and instructions
- Compiled the qft tools
I have calculated the one-loop QED renormalization constants and verified the Ward identity using the computer algebra backend. The calculations assume the MS scheme, Feynman gauge, and a metric convention of (+,-,-,-).
QED One-Loop Renormalization Data
| Constant | Symbolic Expression (MS Scheme) |
|---|---|
| $\delta Z_1$ | -alpha/(2*pi*epsilon) |
| $\delta Z_2$ | -alpha/(2*pi*epsilon) |
| $\delta Z_3$ | -2*alpha/(3*pi*epsilon) |
| $\delta m$ | 3*alpha*m/(2*pi*epsilon) |
| Ward Identity | $Z_1 = Z_2$ is True |
| $\beta(e)$ | e**3/(12*pi**2) |
| $\beta(\alpha)$ | 2*alpha**2/(3*pi) |
The results show that the vertex renormalization constant $Z_1$ and the fermion self-energy constant $Z_2$ are identical at one-loop, as required by gauge invariance. The beta function for $\alpha$ is positive, consistent with the expected running of the QED coupling constant.
Next steps
- Use the
master_integraltool if you need to derive the specific pole structure for a custom kinematic point. - Run
gamma_traceto compute the numerator algebra for the vacuum polarization tensor. - Reference
reference/notes.mdto review the underlying integral definitions.
Connects securely to your tools. The creator never sees your data.
About this skill
Git Hub Repo: https://github.com/gopalsharma1615-ai/qft-renorm-ai.
LLMs frequently fail at QED loop calculations, introducing sign errors, factor-of-2 mistakes, and incorrect 1/epsilon poles. This skill offloads the algebra to a SymPy-backed engine to provide exact symbolic results for one-loop renormalization and Dirac algebra.
What it does
- Dirac Algebra performs gamma matrix traces and contractions in 4D or n-dimensions.
- Master Integrals solves dimensionally regularized integrals for various powers and numerators.
- Counterterm Calculation generates exact expressions for Z1, Z2, Z3, and mass counterterms in the MS scheme.
- Series Expansion produces Laurent series in epsilon for expressions involving Gamma functions and other QFT constants.
- Symbolic Verification checks equality between two physics expressions using symbolic simplification and numeric fallbacks.
How it works
- Decompose the physics problem into discrete algebraic steps such as Dirac traces or Feynman parameterization.
- Execute the specific CLI tool for the step, providing arguments like indices, numerators, or expressions in JSON format.
- Analyze the returned JSON which contains exact symbolic strings or boolean verification results.
- Finalize the result by checking the final expression against textbook conventions using the
checkutility.
Frameworks & tools
Built on SymPy for symbolic math, with NumPy and SciPy for numeric verification. Interfaces via a Python-based CLI wrapper for the qft-renorm-ai backend.
Why this beats prompting it yourself
Zero-shot LLM prompts often hallucinate the coefficients of the QED beta function or drop factors of $i$ in master integrals. This skill uses a deterministic computer algebra system to ensure every sign and factor of $2\pi$ is physically accurate.
Use cases
- Calculating the one-loop vertex correction trace.
- Verifying the Ward Identity ($Z_1 = Z_2$) for QED.
- Expanding dimensionally regularized amplitudes near $n=4$.
- Deriving the QED beta function from first principles.
Known limitations
Supports one-loop QED only. No support for QCD color algebra, $\gamma^5$ matrices, or multi-loop integrals. Requires the qft-renorm-ai package installed in the environment.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Research & Analysis.
One-loop QED only. No QCD color algebra, gamma5, or two-loop support. Requires local installation of the qft-renorm-ai package.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 6 days ago
- Free with an Agensi account
- Software
Python