Algis DumbrisandClaude Opus 4.6
8df22457ab
Trust scores: per (agent, action_type) pair, stored in agent_trust
table. Auto-adjusts when human reacts to AI agent messages (approve
+0.05, reject -0.1). Scores clamped [0.0, 1.0]. MCP get_trust action
+ REST API /api/trust/{agent}. Web UI shows trust progress bars on
agent detail pages.
Claim semantics: only one in_progress reaction per message enforced.
First agent to claim wins, duplicates rejected with clear error.
State-change webhooks: StateChangeNotifier interface fires
workflow.state_changed events through existing webhook infrastructure
when reactions change a message's derived workflow state.
Channel thresholds: publish_threshold and approve_threshold fields
on channels for configuring autonomy gates.
Migration 014_trust_claims.sql. 17 new test cases across trust
model + store.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
18 lines
671 B
SQL
18 lines
671 B
SQL
-- Trust scores per (agent, action_type) for graduated autonomy
|
|
CREATE TABLE IF NOT EXISTS agent_trust (
|
|
id INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
agent_name TEXT NOT NULL,
|
|
action_type TEXT NOT NULL,
|
|
score REAL NOT NULL DEFAULT 0.0,
|
|
adjustments_count INTEGER NOT NULL DEFAULT 0,
|
|
last_adjusted_at TIMESTAMP,
|
|
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
|
|
UNIQUE(agent_name, action_type)
|
|
);
|
|
|
|
CREATE INDEX idx_trust_agent ON agent_trust(agent_name);
|
|
|
|
-- Channel autonomy thresholds
|
|
ALTER TABLE channels ADD COLUMN publish_threshold REAL NOT NULL DEFAULT 0.8;
|
|
ALTER TABLE channels ADD COLUMN approve_threshold REAL NOT NULL DEFAULT 0.6;
|