feat: add 'mem learn' CLI for markdown knowledge ingestion

6 knowledge files: rust, SOLID/DRY, ast-grep, karpathy, golang, caveman
65 chunks ingested to log/knowledge/learn/latest.jsonl
Chunks on ## headings, SHA256 dedup, configurable chunk size

crates/mem-cli/Cargo.toml

@@ -40,3 +40,4 @@ jsonwebtoken = { workspace = true }
 reqwest = { workspace = true }
 async-trait = { workspace = true }
 urlencoding = { workspace = true }
+walkdir = "2.5"

crates/mem-cli/src/main.rs

@@ -155,6 +155,25 @@ enum Commands {
         #[arg(long, value_name = "FILE")]
         file: Option<PathBuf>,
     },
+
+    /// Ingest markdown knowledge files into memory
+    Learn {
+        /// Markdown files or directories to ingest
+        #[arg(value_name = "PATH")]
+        paths: Vec<PathBuf>,
+
+        /// Project to file under
+        #[arg(long, default_value = "knowledge")]
+        project: String,
+
+        /// Dry run — show chunks without writing
+        #[arg(long)]
+        dry_run: bool,
+
+        /// Maximum chunk size in characters (splits on headings)
+        #[arg(long, default_value_t = 2000)]
+        chunk_size: usize,
+    },
 }
 
 #[tokio::main]
@@ -213,6 +232,9 @@ async fn main() -> anyhow::Result<()> {
         Commands::Sig { tool, file } => {
             cmd_sig(&tool, file.as_ref())?
         }
+        Commands::Learn { paths, project, dry_run, chunk_size } => {
+            cmd_learn(&paths, &project, dry_run, chunk_size)?;
+        }
     }
 
     Ok(())
@@ -379,6 +401,150 @@ async fn cmd_verify(
     Ok(())
 }
 
+fn cmd_learn(
+    paths: &[PathBuf],
+    project: &str,
+    dry_run: bool,
+    max_chunk: usize,
+) -> anyhow::Result<()> {
+    use sha2::{Digest, Sha256};
+
+    let mut all_files: Vec<PathBuf> = Vec::new();
+    for p in paths {
+        if p.is_dir() {
+            for entry in walkdir::WalkDir::new(p)
+                .into_iter()
+                .filter_map(|e| e.ok())
+                .filter(|e| {
+                    e.path()
+                        .extension()
+                        .map(|ext| ext == "md")
+                        .unwrap_or(false)
+                })
+            {
+                all_files.push(entry.into_path());
+            }
+        } else if p.extension().map(|e| e == "md").unwrap_or(false) {
+            all_files.push(p.clone());
+        } else {
+            eprintln!("Skipping non-markdown file: {}", p.display());
+        }
+    }
+
+    if all_files.is_empty() {
+        eprintln!("No markdown files found.");
+        return Ok(());
+    }
+
+    all_files.sort();
+    println!("Found {} markdown files", all_files.len());
+
+    let mut total_chunks = 0usize;
+    let mut total_bytes = 0usize;
+    let mut log = if !dry_run {
+        Some(mem_store::LogWriter::new(project, "learn", "latest")?)
+    } else {
+        None
+    };
+
+    for file in &all_files {
+        let content = fs::read_to_string(file)?;
+        let filename = file.file_stem().unwrap().to_string_lossy();
+        let chunks = chunk_markdown(&content, max_chunk);
+
+        println!("\n📄 {} — {} chunks", file.display(), chunks.len());
+
+        for (i, chunk) in chunks.iter().enumerate() {
+            let mut hasher = Sha256::new();
+            hasher.update(chunk.as_bytes());
+            let hash = format!("{:x}", hasher.finalize());
+            let short_hash = &hash[..12];
+
+            total_chunks += 1;
+            total_bytes += chunk.len();
+
+            if dry_run {
+                let preview: String = chunk.chars().take(80).collect();
+                println!(
+                    "  [{}/{}] {} ({} bytes) {}",
+                    i + 1,
+                    chunks.len(),
+                    short_hash,
+                    chunk.len(),
+                    preview.replace('\n', " ")
+                );
+            } else {
+                let record = mem_store::EventRecord {
+                    project: project.to_string(),
+                    query: format!("{}:{}", filename, i),
+                    run: "latest".to_string(),
+                    turn: i as u32,
+                    event_type: "learn".to_string(),
+                    data: serde_json::json!({
+                        "source": file.to_string_lossy(),
+                        "chunk_index": i,
+                        "total_chunks": chunks.len(),
+                        "sha256": hash,
+                        "level": "L1",
+                        "text": chunk,
+                    }),
+                };
+                log.as_mut().unwrap().log(record)?;
+                println!("  ✓ [{}/{}] {} ({} bytes)", i + 1, chunks.len(), short_hash, chunk.len());
+            }
+        }
+    }
+
+    println!("\n{}", "─".repeat(50));
+    println!(
+        "{} files → {} chunks ({:.1} KB)",
+        all_files.len(),
+        total_chunks,
+        total_bytes as f64 / 1024.0
+    );
+    if dry_run {
+        println!("(dry run — nothing written)");
+    } else {
+        println!("Written to log/{}/learn/latest.jsonl", project);
+    }
+    Ok(())
+}
+
+/// Split markdown on ## headings, respecting max_chunk size.
+fn chunk_markdown(content: &str, max_chunk: usize) -> Vec<String> {
+    let mut chunks = Vec::new();
+    let mut current = String::new();
+
+    for line in content.lines() {
+        // Split on ## headings (keep # title in first chunk)
+        if line.starts_with("## ") && !current.is_empty() {
+            let trimmed = current.trim().to_string();
+            if !trimmed.is_empty() {
+                chunks.push(trimmed);
+            }
+            current = String::new();
+        }
+
+        current.push_str(line);
+        current.push('\n');
+
+        // Hard split if chunk too large
+        if current.len() > max_chunk {
+            let trimmed = current.trim().to_string();
+            if !trimmed.is_empty() {
+                chunks.push(trimmed);
+            }
+            current = String::new();
+        }
+    }
+
+    let trimmed = current.trim().to_string();
+    if !trimmed.is_empty() {
+        chunks.push(trimmed);
+    }
+    chunks
+}
+
 fn cmd_sig(tool: &str, file: Option<&PathBuf>) -> anyhow::Result<()> {
     use mem_core::lesson;
     use std::io::Read;