flame_chase
Put two agents on one task and let them take turns. Each turn is a fresh session that starts from the task and the repository, so the working tree is all that passes between them, and every turn goes to an agent that did not write what it is looking at. Two different models fail differently, which suits long work where one agent's blind spots would compound. The name comes from the Flame-Chase Journey (逐火之旅) in Honkai: Star Rail, taken up anew each time its world goes round again.
In hmz, open /flow → Flowverses → official and install flame_chase. Or run the release without installing anything:
hmz exec -f 'git+https://github.com/humanfia/flame-chase-flow@v0.1.0#flame_chase' …- repository
- humanfia/flame-chase-flow
- release
- v0.1.0 · commit
420e349 - licence
- Apache-2.0
Read from humanfia/flowverse when this site was built.
The whole run, at rest. Step through it with the buttons, or drag the bar.
Neither is told what the other said: what passes between them is the working directory. A turn that fails ends the run, and --resume picks up with the chaser whose turn was next.
The run, turn by turn
- first_chaser — reads the CSVs; a session opened for this turn; hands second_chaser in the tree: CSV passes
- second_chaser — adds JSON input; a session opened for this turn; hands first_chaser in the tree: CSV and JSON pass
- first_chaser — adds XML input; a session opened for this turn; hands second_chaser in the tree: XML encoding breaks
- second_chaser — fixes XML encoding; a session opened for this turn; hands first_chaser in the tree: all but .xlsx
- first_chaser — adds .xlsx; a session opened for this turn
Then round again: a round is a turn each.
It ends when the budget runs out, or a turn fails.
See also: flame_chasoid:agent_cleanup_flame_chase · flame_chasoid:parallel · rlar