Close the doc gaps the audit found in the agent knowledge base and the human
walkthrough:
- config-reference: add the Grok provider env table (host ~/.grok subscription
mount, grok-build, idle-kill, cost cap) and the Self-Healing CI loop toggles.
- agent-model: provider-aware Model Configuration (ANTHROPIC default / GROK) +
add the pr_reviewer / prompter / secretary roles to the Roles table.
- tool-permissions: 'three' -> five MCP servers (roboco-optimal, roboco-docs) +
PR Reviewer / Prompter / Secretary tool sections.
- new roles/pr-reviewer.md (the 22nd agent had no role doc); permissions +
agent-uuids + task-tools 'PR Reviewer flow' all gain the role.
- api-endpoints: drop the removed USAGE_UPDATE event (only USAGE_SNAPSHOT exists).
- how-to: self-healing CI loop + Company Scorecard (ch.5), inbound external-PR
review + CEO Supersede/Dismiss queue (ch.4).
Every claim verified against current code by the audit (grok model grok-build,
auth ~/.grok, no metered API; opencode fully removed).
Markdown and editors soft-wrap on their own, so the manual ~75-char line
breaks across the docs added nothing but noise. Join wrapped prose, list
items, and paragraphs into single lines across 67 docs — README, CLAUDE.md,
deployment, usage, the RAG knowledge base, and the agent role prompts.
Whitespace-only: code fences, tables, and blockquote alerts are byte-identical
and the change is token-verified (no content altered). Applied with a
deterministic reflow tool (committed separately).
Also lands two doc edits that were awaiting commit: the measured under-load
resource numbers in usage.md and the pr_reviewer additions to the
org-structure RAG doc.
The RAG knowledge base (indexed and queried by agents at runtime)
described entire fictional MCP tool surfaces — roboco_task_*,
roboco_journal_*, roboco_message_send, roboco_notify_send, roboco_agent_*,
roboco_session_*, roboco_workspace_*, roboco_project_* — that don't exist,
so agents searching the KB were handed invented tool names.
Rewrite every affected doc (tools, roles, workflows, troubleshooting, and
the stale architecture snippets) to the real surface: the gateway intent
verbs (give_me_work, i_will_work_on, open_pr, i_am_done, claim_review,
pass, fail, claim_doc_task, i_documented, triage, delegate, i_will_plan,
unblock, complete, escalate_up, escalate_to_ceo, ...) and content tools
(commit, note(scope=...), say, dm, evidence, notify*, open_session,
channels). Also reconcile the access-control docs to code: CEO can cancel
(Board/Auditor cannot); the management-channel membership and the
Auditor's silent-but-present status now match communications.py.
Renn F