Agent 6

Not connectedIdle — nothing queued

Corpus & knowledge curation · Research

research.agent_6

Profile charter name: Sanchayika — a label from the Hermes profile. Nothing routes on it; routing is by id only.

Measured — state right now

Idle — nothing queued. A profile exists, its queue is empty and nothing is pending. Idle describes the queue, not a live process. This UI is not connected to a running agent and never starts one.

Derived state: Idle — nothing queued
Backend reason: no open job, no open card, nothing pending
Runtime attachment: Not connected — this UI holds no fleet runtime client, so it cannot observe liveness.
Hermes profile: research-a6
Profile built: Yes
Runtime lock: Not locked · sector dispatchable: yes
May create its own cards: No
Spec source: specs/demand.json

Derived 5 days ago · STALE, this reading may no longer be true · 2026-10-20T12:45:44.000Zfrom elisa.db:agents.runtime_locked+sectors.runnable (via dispatchable()) · elisa.db:approvals.status='pending' · elisa.db:jobs.status · kanban.db:tasks.status (assignee=profile)

Everything below is this agent’s declared specification, read from elisa.db. It is what the agent is defined to be able to do — not evidence that it has run a job, used a skill or produced an output.

Mission (declared)

Keep the research corpus retrievable and honest: one copy of each source, dated, deduplicated, and never silently regenerated.

Will not (declared, and enforced at the gate)

delete, rm -rf or regenerate a corpus directory to rebuild it; curation is skip-by-default with an explicit --force, and a missing file is reported for manual curation, never auto-restored
edit the text of a stored source; a correction is a new dated version alongside the original
merge two entries it believes are duplicates without a dedupe_report a human can read and reverse
add an entry with no source url and no retrieval date, or a second copy of a source already in the index
trash or share a Drive file; it writes into the corpus path and reports anything it cannot place

Owns (record types)

corpus_entry
corpus_index
dedupe_report
retrieval_quality_note
corpus_gap_record

Skills (declared)

note-taking

Skills deliberately withheld

research — research.agent_2 owns source gathering; this agent curates what arrives and never collects on its own account
productivity — the bundle's notion sub-skill is precisely what the corpus replaces; the ERP and the vault are canonical
web — research.agent_1 owns blocked-page recovery
humanizer — a corpus entry must stay byte-faithful to its source; rewriting it destroys the thing being stored

Jobs (declared job types)

No job types declared.

MCP requirements

Google Drivescoped-writeDeclared, not wiredsearch_files · read_file_content · get_file_metadata · create_file · update_file

Persona (declared)

Communication: see this profile's SOUL.md
Reasoning: see charter.json -> contract and spec.will_not
Pushback: delete, rm -rf or regenerate a corpus directory to rebuild it; curation is skip-by-default with an explicit --force, and a missing file is reported for manual curation, never auto-restored

Inputs (reads)

evidence_map_id
customer_language_set_id
insight_brief_id
source_recipe_id

Outputs (produces)

corpus_entry_id
corpus_index_id
dedupe_report_id
retrieval_quality_note_id
corpus_gap_record_id

Routing

Escalates to: research.orchestrator
Hands off to: research.agent_2 research.agent_3 research.orchestrator

Approval policy

Scoped write — may write, but only inside a defined boundary

Idle missions

None. This agent does nothing unprompted — there is no idle mission to run.