Remove external Example run references from README.md and AGENTS.md

Agent-Logs-Url: https://github.com/githubnext/autoloop/sessions/fcb97875-6be2-4fec-af02-6994796487df

Co-authored-by: mrjf <180956+mrjf@users.noreply.github.com>

Author: Copilot

AGENTS.md

@@ -80,9 +80,7 @@ Programs can include an Evolution Strategy section (inspired by OpenEvolve) that
 
 - **Agentic Workflows**: https://github.com/github/gh-aw
 - **Quick Start**: https://github.github.com/gh-aw/setup/quick-start/
-- **Autoloop Examples**: https://github.com/githubnext/autoloop_examples
-  - Function minimization: [code](https://github.com/githubnext/autoloop_examples/tree/main/examples/function_minimization), [actions run](https://github.com/githubnext/autoloop_examples/actions/runs/23503588936), [PR](https://github.com/githubnext/autoloop_examples/pull/4)
-  - Signal processing: [code](https://github.com/githubnext/autoloop_examples/tree/main/examples/signal_processing), [actions run](https://github.com/githubnext/autoloop_examples/actions/runs/23511458560), [PR](https://github.com/githubnext/autoloop_examples/pull/11)
+- **Autoloop Examples**: See the [example programs](.autoloop/programs/) included in this repo
 
 ## Conventions
 

README.md

@@ -71,23 +71,20 @@ Minimizes validation bits-per-byte (`val_bpb`) for LLM pretraining within a fixe
 
 - **Metric**: `combined_score` = `1/val_bpb` (higher is better)
 - **Code**: [`.autoloop/programs/autoresearch/`](.autoloop/programs/autoresearch/)
-- **Example run**: [actions](https://github.com/githubnext/autoresearch_local/actions/runs/23924736866) | [PR](https://github.com/githubnext/autoresearch_local/pull/14)
 
 ### Function Minimization
 
 Discovers optimization algorithms to find the global minimum of a multi-modal function `f(x,y) = sin(x)*cos(y) + sin(x*y) + (x^2+y^2)/20`. Starts from naive random search and evolves toward techniques like simulated annealing, basin-hopping, and gradient estimation.
 
 - **Metric**: `combined_score` (higher is better), baseline 0.56
 - **Code**: [`.autoloop/programs/function_minimization/`](.autoloop/programs/function_minimization/)
-- **Example run**: [actions](https://github.com/githubnext/autoloop_examples/actions/runs/23503588936) | [PR](https://github.com/githubnext/autoloop_examples/pull/4)
 
 ### Signal Processing
 
 Discovers and optimizes real-time adaptive filtering algorithms for noisy, non-stationary time series. Starts from a simple weighted moving average and evolves toward Kalman filters, wavelet denoising, and hybrid approaches.
 
 - **Metric**: `overall_score` (higher is better), baseline 0.34
 - **Code**: [`.autoloop/programs/signal_processing/`](.autoloop/programs/signal_processing/)
-- **Example run**: [actions](https://github.com/githubnext/autoloop_examples/actions/runs/23511458560) | [PR](https://github.com/githubnext/autoloop_examples/pull/11)
 
 ### Circle Packing