diff --git a/README.md b/README.md
index 8000319..15feffb 100644
--- a/README.md
+++ b/README.md
@@ -12,11 +12,11 @@ Because all identity, communication, and memory live on Nostr, the agent is **po
 
 **Skills are the new apps.** Agents learn capabilities through skills — public Nostr events that any agent can discover, adopt, and share. There is no app store, no gatekeeper, no approval process. If someone publishes a useful skill, your agent can find it through your web of trust and start using it. Popularity is measured by adoption, not by a rating algorithm. The best skills spread because agents actually use them.
 
-## Current Status — v0.0.14
+## Current Status — v0.0.15
 
 **Active build — this project is barely working. Experiment at your own risk.**
 
-> Last release update: v0.0.14 — Handle malformed nostr_post arguments with loose JSON fallback parsing
+> Last release update: v0.0.15 — Auto-add NIP-23 title/image/summary/published_at/d tags when missing
 
 - Connects to configured relays with auto-reconnect and relay state transition logging
 - Publishes configured startup events per relay as each relay becomes connected
diff --git a/src/main.h b/src/main.h
index 697efb0..6e4082a 100644
--- a/src/main.h
+++ b/src/main.h
@@ -12,8 +12,8 @@
 // Using DIDACTYL_ prefix to avoid conflicts with nostr_core_lib VERSION macros
 #define DIDACTYL_VERSION_MAJOR 0
 #define DIDACTYL_VERSION_MINOR 0
-#define DIDACTYL_VERSION_PATCH 14
-#define DIDACTYL_VERSION "v0.0.14"
+#define DIDACTYL_VERSION_PATCH 15
+#define DIDACTYL_VERSION "v0.0.15"
 
 // Agent metadata
 #define DIDACTYL_NAME "Didactyl"
diff --git a/src/tools.c b/src/tools.c
index 11c5cf3..54eb004 100644
--- a/src/tools.c
+++ b/src/tools.c
@@ -7,6 +7,7 @@
 #include <string.h>
 #include <limits.h>
 #include <ctype.h>
+#include <time.h>
 
 #include "cjson/cJSON.h"
 #include "nostr_handler.h"
@@ -262,6 +263,276 @@ static cJSON* parse_loose_nostr_post_args(const char* in) {
     return args;
 }
 
+static cJSON* ensure_tags_array(cJSON** tags_inout) {
+    if (!tags_inout) return NULL;
+
+    if (*tags_inout) {
+        if (cJSON_IsArray(*tags_inout)) {
+            return *tags_inout;
+        }
+        cJSON_Delete(*tags_inout);
+        *tags_inout = NULL;
+    }
+
+    *tags_inout = cJSON_CreateArray();
+    return *tags_inout;
+}
+
+static int has_tag_key(cJSON* tags, const char* key) {
+    if (!tags || !cJSON_IsArray(tags) || !key) return 0;
+
+    int n = cJSON_GetArraySize(tags);
+    for (int i = 0; i < n; i++) {
+        cJSON* tag = cJSON_GetArrayItem(tags, i);
+        if (!tag || !cJSON_IsArray(tag)) continue;
+
+        cJSON* k = cJSON_GetArrayItem(tag, 0);
+        if (k && cJSON_IsString(k) && k->valuestring && strcmp(k->valuestring, key) == 0) {
+            return 1;
+        }
+    }
+    return 0;
+}
+
+static int add_string_tag(cJSON* tags, const char* key, const char* value) {
+    if (!tags || !cJSON_IsArray(tags) || !key || !value || value[0] == '\0') return -1;
+
+    cJSON* tag = cJSON_CreateArray();
+    if (!tag) return -1;
+
+    cJSON_AddItemToArray(tag, cJSON_CreateString(key));
+    cJSON_AddItemToArray(tag, cJSON_CreateString(value));
+    cJSON_AddItemToArray(tags, tag);
+    return 0;
+}
+
+static char* trim_copy(const char* start, size_t len) {
+    while (len > 0 && isspace((unsigned char)start[0])) {
+        start++;
+        len--;
+    }
+    while (len > 0 && isspace((unsigned char)start[len - 1])) {
+        len--;
+    }
+
+    char* out = (char*)malloc(len + 1U);
+    if (!out) return NULL;
+    memcpy(out, start, len);
+    out[len] = '\0';
+    return out;
+}
+
+static char* first_markdown_h1(const char* content) {
+    if (!content) return NULL;
+
+    const char* p = content;
+    while (*p) {
+        const char* line = p;
+        const char* nl = strchr(line, '\n');
+        size_t len = nl ? (size_t)(nl - line) : strlen(line);
+
+        while (len > 0 && (line[len - 1] == '\r')) len--;
+
+        size_t i = 0;
+        while (i < len && isspace((unsigned char)line[i])) i++;
+        if (i < len && line[i] == '#') {
+            size_t j = i;
+            while (j < len && line[j] == '#') j++;
+            if (j < len && isspace((unsigned char)line[j])) {
+                while (j < len && isspace((unsigned char)line[j])) j++;
+                if (j < len) return trim_copy(line + j, len - j);
+            }
+        }
+
+        if (!nl) break;
+        p = nl + 1;
+    }
+    return NULL;
+}
+
+static int is_paragraph_line(const char* line, size_t len) {
+    size_t i = 0;
+    while (i < len && isspace((unsigned char)line[i])) i++;
+    if (i >= len) return 0;
+
+    char c = line[i];
+    if (c == '#' || c == '>' || c == '-' || c == '*' || c == '`' || c == '|') return 0;
+    if (isdigit((unsigned char)c)) {
+        size_t j = i;
+        while (j < len && isdigit((unsigned char)line[j])) j++;
+        if (j < len && line[j] == '.') return 0;
+    }
+    return 1;
+}
+
+static char* first_markdown_paragraph(const char* content) {
+    if (!content) return NULL;
+
+    const char* p = content;
+    while (*p) {
+        const char* line = p;
+        const char* nl = strchr(line, '\n');
+        size_t len = nl ? (size_t)(nl - line) : strlen(line);
+        while (len > 0 && line[len - 1] == '\r') len--;
+
+        if (is_paragraph_line(line, len)) {
+            size_t cap = 1024;
+            size_t used = 0;
+            char* out = (char*)malloc(cap);
+            if (!out) return NULL;
+
+            const char* q = line;
+            const char* qnl = nl;
+            size_t qlen = len;
+
+            while (1) {
+                if (!is_paragraph_line(q, qlen)) break;
+
+                size_t start = 0;
+                while (start < qlen && isspace((unsigned char)q[start])) start++;
+                size_t end = qlen;
+                while (end > start && isspace((unsigned char)q[end - 1])) end--;
+                size_t part_len = end - start;
+
+                if (used + part_len + 2 >= cap) {
+                    cap = (cap * 2U) + part_len + 16U;
+                    char* bigger = (char*)realloc(out, cap);
+                    if (!bigger) {
+                        free(out);
+                        return NULL;
+                    }
+                    out = bigger;
+                }
+
+                if (part_len > 0) {
+                    if (used > 0) out[used++] = ' ';
+                    memcpy(out + used, q + start, part_len);
+                    used += part_len;
+                }
+
+                if (!qnl) break;
+                q = qnl + 1;
+                qnl = strchr(q, '\n');
+                qlen = qnl ? (size_t)(qnl - q) : strlen(q);
+                while (qlen > 0 && q[qlen - 1] == '\r') qlen--;
+            }
+
+            out[used] = '\0';
+            return out;
+        }
+
+        if (!nl) break;
+        p = nl + 1;
+    }
+
+    return NULL;
+}
+
+static char* first_markdown_image_url(const char* content) {
+    if (!content) return NULL;
+
+    const char* p = content;
+    while ((p = strstr(p, "![")) != NULL) {
+        const char* close_bracket = strchr(p + 2, ']');
+        if (!close_bracket || close_bracket[1] != '(') {
+            p += 2;
+            continue;
+        }
+
+        const char* url_start = close_bracket + 2;
+        const char* url_end = strchr(url_start, ')');
+        if (!url_end || url_end <= url_start) {
+            p += 2;
+            continue;
+        }
+
+        return trim_copy(url_start, (size_t)(url_end - url_start));
+    }
+
+    return NULL;
+}
+
+static char* slugify_string(const char* in, const char* fallback) {
+    if (!in || in[0] == '\0') {
+        return fallback ? strdup(fallback) : NULL;
+    }
+
+    size_t len = strlen(in);
+    char* out = (char*)malloc(len + 1U);
+    if (!out) return NULL;
+
+    size_t j = 0;
+    int prev_dash = 0;
+    for (size_t i = 0; i < len; i++) {
+        unsigned char c = (unsigned char)in[i];
+        if (isalnum(c)) {
+            out[j++] = (char)tolower(c);
+            prev_dash = 0;
+        } else if (!prev_dash && j > 0) {
+            out[j++] = '-';
+            prev_dash = 1;
+        }
+    }
+
+    while (j > 0 && out[j - 1] == '-') j--;
+    out[j] = '\0';
+
+    if (j == 0) {
+        free(out);
+        return fallback ? strdup(fallback) : NULL;
+    }
+
+    return out;
+}
+
+static void ensure_nip23_metadata_tags(int kind, const char* content, cJSON** tags_inout) {
+    if (!content || (kind != 30023 && kind != 30024) || !tags_inout) {
+        return;
+    }
+
+    cJSON* tags = ensure_tags_array(tags_inout);
+    if (!tags) return;
+
+    char* title = has_tag_key(tags, "title") ? NULL : first_markdown_h1(content);
+    char* summary = has_tag_key(tags, "summary") ? NULL : first_markdown_paragraph(content);
+    char* image = has_tag_key(tags, "image") ? NULL : first_markdown_image_url(content);
+
+    if (title) {
+        (void)add_string_tag(tags, "title", title);
+    }
+    if (summary) {
+        (void)add_string_tag(tags, "summary", summary);
+    }
+    if (image) {
+        (void)add_string_tag(tags, "image", image);
+    }
+
+    if (!has_tag_key(tags, "published_at")) {
+        char published_at[32];
+        snprintf(published_at, sizeof(published_at), "%lld", (long long)time(NULL));
+        (void)add_string_tag(tags, "published_at", published_at);
+    }
+
+    if (!has_tag_key(tags, "d")) {
+        char* slug = NULL;
+        if (title) {
+            slug = slugify_string(title, "long-form-note");
+        } else {
+            char* derived_title = first_markdown_h1(content);
+            slug = slugify_string(derived_title, "long-form-note");
+            free(derived_title);
+        }
+        if (slug) {
+            (void)add_string_tag(tags, "d", slug);
+            free(slug);
+        }
+    }
+
+    free(title);
+    free(summary);
+    free(image);
+}
+
 static int is_safe_relative_path(const char* path) {
     if (!path || path[0] == '\0') return 0;
     if (path[0] == '/') return 0;
@@ -487,6 +758,8 @@ static char* execute_nostr_post(const char* args_json) {
         }
     }
 
+    ensure_nip23_metadata_tags((int)kind->valuedouble, content->valuestring, &tags_dup);
+
     nostr_publish_result_t publish_result;
     memset(&publish_result, 0, sizeof(publish_result));