Teknium
5127567d5d
perf(ttft): cache skills prompt with shared skill_utils module (salvage #3366) (#3421)
Two-layer caching for build_skills_system_prompt():
1. In-process LRU (OrderedDict, max 8) — same-process: 546ms → <1ms
2. Disk snapshot (.skills_prompt_snapshot.json) — cold start: 297ms → 103ms
Key improvements over original PR #3366:
- Extract shared logic into agent/skill_utils.py (parse_frontmatter,
skill_matches_platform, get_disabled_skill_names, extract_skill_conditions,
extract_skill_description, iter_skill_index_files)
- tools/skills_tool.py delegates to shared module — zero code duplication
- Proper LRU eviction via OrderedDict.move_to_end + popitem(last=False)
- Cache invalidation on all skill mutation paths:
- skill_manage tool (in-conversation writes)
- hermes skills install (CLI hub)
- hermes skills uninstall (CLI hub)
- Automatic via mtime/size manifest on cold start
prompt_builder.py no longer imports tools.skills_tool (avoids pulling
in the entire tool registry chain at prompt build time).
6301 tests pass, 0 failures.
Co-authored-by: kshitijk4poor <82637225+kshitijk4poor@users.noreply.github.com>
2026-03-27 10:54:02 -07:00
..
2026-02-21 22:31:43 -08:00
2026-03-27 07:49:44 -07:00
2026-03-27 09:45:25 -07:00
2026-03-25 15:02:03 -07:00
2026-03-23 06:40:05 -07:00
2026-03-20 09:38:13 -07:00
2026-03-26 17:40:31 -07:00
2026-03-25 19:47:58 -07:00
2026-03-26 02:00:50 -07:00
2026-03-20 08:52:37 -07:00
2026-03-27 10:54:02 -07:00
2026-03-21 16:54:43 -07:00
2026-03-21 16:55:02 -07:00
2026-03-18 03:17:37 -07:00
2026-03-27 10:54:02 -07:00
2026-03-17 23:40:22 -07:00
2026-03-17 04:14:40 -07:00
2026-02-21 22:31:43 -08:00
2026-03-25 12:45:58 -07:00