Restore load_skill + surface skill catalog in system prompt

Main's load_skill tool was deleted during the SDK migration along
with the prompt-mutation pattern it relied on. Re-add the capability
without the mutation: load_skill(skills=[...]) now returns the skill
markdown bodies as a tool result, so the content lands in conversation
history as in-context reference rather than as patched-in system
prompt content. Same source of truth (load_skills + skill files),
same validation (validate_requested_skills) as create_agent.

Tool result format is plain markdown (## Skill: <name> headers joined
with ---), not the <specialized_knowledge> XML wrapping used at
agent-build time. The XML framing was deliberately reserved for
prompt-level privileged context; tool-loaded skills are honestly
labelled as just-fetched reference material.

Close the discovery loop by surfacing the full skill catalog in the
system prompt. Without it the model could only guess skill names —
discovering them via validation errors on misses. Now every agent
sees a categorised <available_skills> block right after the
<specialized_knowledge> block with a short hint pointing at
create_agent / load_skill.

Skills module: factored _iter_user_skill_files() so get_all_skill_names
(set, for validation) and get_available_skills (dict by category,
for the prompt) share one source of truth on what counts as
user-selectable. Internal categories (scan_modes, coordination) stay
excluded from both.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

strix/tools/load_skill/tool.py

@@ -0,0 +1,36 @@
+"""``load_skill`` — fetch skill reference material into the conversation."""
+
+from __future__ import annotations
+
+from agents import RunContextWrapper, function_tool
+
+from strix.skills import load_skills, validate_requested_skills
+
+
+@function_tool(timeout=10)
+async def load_skill(ctx: RunContextWrapper, skills: list[str]) -> str:
+    """Return the markdown body of one or more skills as reference material.
+
+    Use this when you need exact syntax / workflow / payload guidance
+    right before acting on a technology that wasn't preloaded for your
+    agent. The skill content lands inline as a tool result — no
+    permanent prompt change, just in-conversation reference.
+
+    For permanent skill assignment, pass ``skills=[…]`` to
+    ``create_agent`` when spawning a specialist child instead.
+
+    Args:
+        skills: List of skill names (e.g. ``["xss", "sql_injection"]``).
+            Max 5. Names match the bare files under
+            ``strix/skills/<category>/<name>.md``.
+    """
+    del ctx
+    requested = list(skills or [])
+    err = validate_requested_skills(requested)
+    if err:
+        return f"load_skill: {err}"
+    contents = load_skills(requested)
+    if not contents:
+        return "load_skill: no content loaded for requested skills."
+    sections = [f"## Skill: {name}\n\n{body}" for name, body in contents.items()]
+    return "\n\n---\n\n".join(sections)