hermes-agent/website/docs/user-guide/features
Teknium abf1af5401
feat(session_search): single-shape tool with discovery, scroll, browse — no LLM (#27590)
* feat(session_search): single-shape tool with discovery, scroll, browse — no LLM

Replaces the LLM-summarized session_search with a single-shape tool that
returns actual messages from the DB. Three calling shapes inferred from
args (no mode parameter):

  1. Discovery — pass query. FTS5 + anchored ±5 window + bookends per hit,
     all in one call. ~20ms on a real DB instead of ~90s for the previous
     three aux-LLM calls.
  2. Scroll — pass session_id + around_message_id. Returns a window
     centered on the anchor. To paginate, re-anchor on the first/last id
     of the returned window. Boundary message appears in both windows
     as the orientation marker. ~1ms per scroll call.
  3. Browse — no args. Recent sessions chronologically.

Bookend_start (first 3 user+assistant msgs) and bookend_end (last 3) give
the agent goal + resolution on every discovery hit, so a single tool call
reconstructs a long session's arc without loading the whole transcript.

The aux-LLM summary path is gone: it cost ~$0.30/call, took ~30s, and
laundered FTS5 hits through a model that could confabulate when the right
session wasn't in the hit list. The merged shape returns byte-for-byte
content from SQLite.

History:
- PR #20238 (JabberELF) seeded the fast/summary dual-mode split.
- PR #26419 (yoniebans) expanded to fast/guided/summary with bookends,
  multi-anchor drill-down, default-mode config, and a teaching skill.

This PR collapses that toolkit into one shape with explicit scroll
support, drops the summary path, drops the mode parameter, drops the
config knob, drops the skill. JabberELF's seed work is acknowledged via
the AUTHOR_MAP entry.

Validation:
- 38/38 tool tests pass (tests/tools/test_session_search.py)
- 12/12 get_messages_around tests pass (tests/hermes_state/)
- 11/11 get_anchored_view tests pass (tests/hermes_state/)
- Full tests/tools/ run: 5168 passing, 2 failures pre-exist on main
  (test ordering in test_delegate.py, unrelated)
- E2E against live state DB: discovery 20ms, scroll 1ms, browse 280ms;
  pagination forward+backward works with boundary-message orientation;
  error paths return clean tool_error responses

Co-authored-by: JabberELF <abcdjmm970703@gmail.com>
Co-authored-by: yoniebans <jonny@nousresearch.com>

* chore(session_search): prune dead LLM-summary config and docs

Companion to the single-shape rewrite. The auxiliary.session_search config
block, max_concurrency / extra_body tunables, and matching docs sections
all referenced the removed LLM summarization path. Removing them so users
don't try to tune knobs that nothing reads.

- hermes_cli/config.py: drop dead auxiliary.session_search block from
  DEFAULT_CONFIG. Leftover keys in user config.yaml are harmless and
  ignored.
- hermes_cli/tips.py: drop two tips referencing the removed
  max_concurrency / extra_body knobs.
- website/docs/user-guide/configuration.md: drop 'Session Search Tuning'
  section and the auxiliary.session_search block from the example.
- website/docs/user-guide/features/fallback-providers.md: drop session_search
  rows from the auxiliary-tasks tables and the dedicated tuning subsection.
- website/docs/reference/tools-reference.md: rewrite the session_search
  entry to describe the new three-shape behaviour.
- CONTRIBUTING.md: update the file-tree description.
- tests/tools/test_llm_content_none_guard.py: remove TestSessionSearchContentNone
  class and test_session_search_tool_guarded — both guard against an
  unguarded .content.strip() call site in _summarize_session() that no
  longer exists.

Validation: 97/97 targeted tests still pass (hermes_state + session_search +
llm_content_none_guard). Config tests 55/55.

---------

Co-authored-by: JabberELF <abcdjmm970703@gmail.com>
Co-authored-by: yoniebans <jonny@nousresearch.com>
2026-05-17 23:28:45 -07:00
..
_category_.json feat: add documentation website (Docusaurus) 2026-03-05 05:24:55 -08:00
acp.md feat(acp): hermes acp --setup-browser bootstraps browser tools for registry installs 2026-05-15 01:38:24 -07:00
api-server.md docs: deep audit — fix stale config keys, missing commands, and registry drift (#22784) 2026-05-09 13:19:51 -07:00
batch-processing.md fix: normalize remaining reasoning effort orderings and add missing 'minimal' 2026-04-09 14:20:16 -07:00
browser.md fix(browser): honor pre-set AGENT_BROWSER_ARGS and document the bypass 2026-05-14 19:02:17 -07:00
built-in-plugins.md fix(plugins): remove unreachable hermes tools → Langfuse path 2026-05-16 17:15:19 -07:00
code-execution.md docs(execute_code): document project/strict execution modes (#12073) 2026-04-18 01:53:09 -07:00
codex-app-server-runtime.md fix(codex): allow kanban worker board writes 2026-05-17 11:50:43 -07:00
computer-use.md feat(computer-use): refresh cua-driver on hermes update + add install --upgrade (#24063) 2026-05-11 17:10:58 -07:00
context-files.md feat: progressive subdirectory hint discovery (#5291) 2026-04-05 12:33:47 -07:00
context-references.md docs: comprehensive documentation audit — fix stale info, expand thin pages, add depth (#5393) 2026-04-05 19:45:50 -07:00
credential-pools.md docs(anthropic): correct OAuth scope to Max plan + extra usage credits only (#17404) 2026-04-29 04:11:14 -07:00
cron.md docs(cron): document name-based job lookup from #26231 2026-05-16 20:32:43 -07:00
curator.md docs(curator): update CLI docs for synchronous-by-default manual run 2026-05-07 05:27:47 -07:00
delegation.md docs(delegation): show api_mode override in custom-endpoint example 2026-05-16 20:32:43 -07:00
extending-the-dashboard.md feat(plugins): run any LLM call from inside a plugin via ctx.llm (#23194) 2026-05-10 07:09:28 -07:00
fallback-providers.md feat(session_search): single-shape tool with discovery, scroll, browse — no LLM (#27590) 2026-05-17 23:28:45 -07:00
goals.md makes the Persistent Goals docs accessible in the docs nav (and llms.txt) (#18481) 2026-05-01 10:29:22 -07:00
honcho.md docs: deep audit — fix stale config keys, missing commands, and registry drift (#22784) 2026-05-09 13:19:51 -07:00
hooks.md test+docs: cover transform_llm_output hook + release author map 2026-05-07 05:46:05 -07:00
image-generation.md feat(image-gen): add GPT Image 2 to FAL catalog (#13677) 2026-04-21 13:35:31 -07:00
kanban-tutorial.md feat(kanban): orchestrator-driven auto-decomposition on triage (#27572) 2026-05-17 13:54:12 -07:00
kanban-worker-lanes.md feat(kanban): stranded_in_ready diagnostic for unclaimed tasks (#23578) 2026-05-10 21:58:44 -07:00
kanban.md feat(kanban): orchestrator-driven auto-decomposition on triage (#27572) 2026-05-17 13:54:12 -07:00
lsp.md docs(lsp): replace "git worktree" with "git repository" in LSP docs 2026-05-13 23:05:20 -07:00
mcp.md feat: add supports_parallel_tool_calls for MCP servers (#26825) 2026-05-16 01:04:28 -07:00
memory-providers.md docs: deep audit — fix stale config keys, missing commands, and registry drift (#22784) 2026-05-09 13:19:51 -07:00
memory.md docs: add Supermemory to memory providers docs, env vars, CLI reference 2026-04-06 22:15:58 -07:00
overview.md feat(tts): add Piper as a native local TTS provider (closes #8508) (#17885) 2026-04-30 02:53:20 -07:00
personality.md docs: document SOUL.md as primary agent identity (#1927) 2026-03-18 04:18:08 -07:00
plugins.md fix(plugins): surface category-namespaced plugins in hermes plugins list 2026-05-16 17:15:19 -07:00
provider-routing.md docs: fallback providers + /background command documentation 2026-03-15 06:24:28 -07:00
skills.md feat(skills-hub): add huggingface/skills as trusted default tap (#2549) 2026-05-15 01:25:33 -07:00
skins.md fix(tui): honor skin highlight colors (#20895) 2026-05-06 14:01:56 -07:00
spotify.md docs(spotify): document Home Assistant speaker routing 2026-05-16 20:32:43 -07:00
subscription-proxy.md feat(proxy): local OpenAI-compatible proxy for OAuth providers (#25969) 2026-05-14 15:40:48 -07:00
tool-gateway.md docs(tool-gateway): rewrite as pitch-first marketing page (#20827) 2026-05-06 13:20:09 -07:00
tools.md docs(tools): add video_generate / video_gen toolset to user-facing tool docs (#27050) 2026-05-16 11:53:13 -07:00
tts.md docs(voice): add Doubao speech integration examples (TTS + STT) 2026-05-05 13:54:33 -07:00
vision.md docs: two-week gap sweep — platforms, CLI, config, TUI, hooks, providers (#17727) 2026-04-29 20:32:37 -07:00
voice-mode.md docs(discord): fix Server Members Intent + SSRC-mapping drift; add /voice join slash Choice 2026-05-05 13:50:43 -07:00
web-dashboard.md docs(dashboard): clarify chat tab tui flag 2026-05-16 20:32:43 -07:00
web-search.md docs(web-search): explain auxiliary-model summarization for web_extract (#23211) 2026-05-10 06:40:23 -07:00
x-search.md feat(x_search): gated X (Twitter) search tool with OAuth-or-API-key auth (#26763) 2026-05-16 00:58:27 -07:00