reflect
The reflect verb (retain/recall/reflect): recall over the vault, then a gateway synthesis pass — one on-demand, query-scoped operation returning a grounded answer with source provenance. mode ‘challenge’ runs the adversarial red-team over the decision-bearing recall instead (the knowledge_challenge core) and returns a coverage object (recalled, decision_matched, filtered_out, kept, dropped, unresolved_evidence_paths). persist: true writes the answer as a derived note under the memory folder’s reflections/ with source_model + chunk provenance (requires write:notes); it applies to mode ‘synthesis’ only and is rejected with mode ‘challenge’. Degrades gracefully: without the inference gateway, recall still returns sources with available: false. The sleep-time half (episode-eligibility evaluator + preference profile) runs via the obsidian-tc reflect CLI command.
| Domain | knowledge |
| Scopes | read:notes |
| Access | read |
| Tags | read-only, domain:knowledge, knowledge, external-network |
| Profile | core, full |
Annotations
Section titled “Annotations”| Annotation | Value |
|---|---|
| Read-only | yes |
| Destructive | no |
| Idempotent | no |
| Accepts an idempotency key | no |
Idempotent is the advisory MCP idempotentHint. Destructive is also yes for a tool that
asks for confirmation only on some calls.
Confirmation (human in the loop)
Section titled “Confirmation (human in the loop)”No human confirmation is required to call this tool.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
vault |
string |
— | Vault id. May be omitted when only one vault is available. length 1–64; pattern ^[a-z0-9_-]+$ |
|
query |
string |
yes | — | length 3–4000 |
mode |
enum |
"synthesis" |
One of: "synthesis", "challenge" |
|
k |
integer |
20 |
range 1–60 | |
scope |
string |
— | length ≥ 1 | |
persist |
boolean |
false |
— | |
citation_style |
enum |
— | How citations read in the answer and any persisted note. “numeric” (default) keeps the [n] markers; “wikilink” renders each [n] as [[path]] from the evidence it numbers, deterministically. An [n] with no matching evidence is left as written and listed in unresolved_citations. Omitted: the vault’s reflect.citationStyle, else numeric. One of: "numeric", "wikilink" |
|
detail |
enum |
— | How much the synthesis says. “concise” (default), “standard”, or “thorough”; a prompt instruction only, the evidence set is unchanged. Omitted: the vault’s reflect.detail, else concise. One of: "concise", "standard", "thorough" |
Output
Section titled “Output”| Field | Type | Always present | Description |
|---|---|---|---|
vault |
string |
yes | — |
mode |
string |
yes | — |
route |
string[] |
yes | — |
available |
boolean |
yes | — |
message |
string |
— | |
answer |
string | null |
— | |
model |
string |
— | |
challenge |
object |
— | |
challenge.verdict |
enum |
with parent | One of: "proceed", "reconsider", "do_not_proceed" |
challenge.summary |
string |
with parent | length ≥ 1 |
challenge.categories |
object[] |
with parent | — |
coverage |
object |
— | |
coverage.recalled |
integer |
with parent | range ≥ 0 |
coverage.decision_matched |
integer |
with parent | range ≥ 0 |
coverage.filtered_out |
integer |
with parent | range ≥ 0 |
coverage.kept |
integer |
with parent | range ≥ 0 |
coverage.dropped |
object |
with parent | — |
coverage.truncated |
integer |
with parent | range ≥ 0 |
coverage.low_coverage |
boolean |
with parent | — |
coverage.unresolved_evidence_paths |
string[] |
— | |
coverage.ungrounded_items_dropped |
integer |
range ≥ 0 | |
sources |
object[] |
yes | — |
sources[].chunk_id |
string |
yes | — |
sources[].path |
string |
yes | — |
sources[].score |
number |
yes | — |
persisted |
object |
— | |
persisted.path |
string |
with parent | — |
excluded_count |
integer |
range ≥ 0 | |
unresolved_citations |
integer[] |
citation_style “wikilink”: [n] markers that match no evidence item, left as written in the answer. Omitted when every marker resolved, and always in numeric mode. |
Related tools
Section titled “Related tools”Other tools in the knowledge domain: add_observation, audit_provenance, close_goal, commit_capture, commit_wiki_page, create_entity, delete_entity, diagnose_retrieval, draft_wiki_page, end_session, enqueue_capture, episode_stats, explain_answer, find_existing_page, gap_report, get_entity, get_session_traces, knowledge_challenge, link_entities, lint_wiki, list_capture_queue, list_contradictions, list_goals, note_quality_report, plur_get, plur_recall, plur_recall_hybrid, plur_similarity_search, query_entity_graph, record_retrieval_feedback, rename_entity, session_bootstrap, set_goal, start_session, unlink_entities, update_observation, vault_context, vault_graph_search, work_episode_chain, work_episodes, work_forget, work_result, work_search.
Back to the tool catalog.