Pick any line to see why it is the way it is: the commit, the pull request and issue it came from, and what the agent was thinking.
| The sidebar asks billing whether it is on, not for the account, so no page waits on the card processor | 1 | -- Planning: an outcome someone wrote, turned by an agent into issues with |
| 2 | -- the order they have to land in, which a person reads before any of it | |
| 3 | -- is opened. | |
| 4 | CREATE TABLE plans ( | |
| 5 | id TEXT PRIMARY KEY, | |
| 6 | repo_id TEXT NOT NULL, | |
| 7 | -- The outcome wanted, as written. | |
| 8 | brief TEXT NOT NULL, | |
| 9 | -- planning, ready, failed or applied. | |
| 10 | status TEXT NOT NULL DEFAULT 'planning', | |
| 11 | -- The agent's account of how it split the work. | |
| 12 | summary TEXT, | |
| 13 | -- JSON array of proposed issues; once applied, each carries its number. | |
| 14 | issues TEXT NOT NULL DEFAULT '[]', | |
| 15 | error TEXT, | |
| 16 | -- SHA-256 of the token the sandbox reports with. | |
| 17 | token_hash TEXT NOT NULL, | |
| 18 | author_id TEXT NOT NULL, | |
| 19 | author_name TEXT NOT NULL, | |
| 20 | created_at TEXT NOT NULL, | |
| 21 | finished_at TEXT | |
| 22 | ); | |
| 23 | CREATE INDEX plans_by_repo ON plans (repo_id, id); | |
| 24 | ||
| 25 | -- The issues that have to be merged before this one is worked on: a JSON | |
| 26 | -- array of issue numbers. | |
| 27 | ALTER TABLE issues ADD COLUMN blocked_by TEXT NOT NULL DEFAULT '[]'; | |
| 28 | -- Set when someone asked for a g1t agent to take the issue as soon as it | |
| 29 | -- can: JSON of who asked, on whose say-so the agent then works. Cleared | |
| 30 | -- when a pull request is opened for it. | |
| 31 | ALTER TABLE issues ADD COLUMN queued_by TEXT; |
This file's history is long; its oldest lines are credited to the oldest commit read.