Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →blueberrycongee/hydra
Use when a task should run through Hydra's Lead-driven workflow for multi-agent orchestration, or when an existing workflow must be inspected or cleaned up.
Passed 8 of 14 test cases with the skill, and 4 of 14 without it.
and 3 more in the full report
6 fixed7 passed either way9 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
We need to quickly run a one-off worker task in an isolated container without creating a full tracked multi-step workflow. What command accomplishes this?
| Checked | Without skill | With skill |
|---|---|---|
| Uses hydra spawn for direct isolated worker execution | Missed | Passed |
The instructions your model reads when the skill loads.
Hydra is a Lead-driven orchestration toolkit. You (the Lead agent) make strategic decisions; Hydra handles operational management (dispatch, retry, health checks, result collection).
Sub-agents output semantic intent (done/needs_rework/replan), not routing
information. Hydra manages the lifecycle; you decide what happens next.
hydra watch is PollForDecisionTask; the Lead is the decider; lead_terminal_id enforces single-decider semantics.dispatch + depends_on + worktree + merge are first-class. Other products (Factory.ai's Droid, Amp, Claude Code subagents) treat parallelism as open research; Hydra makes it the default.result.json (outcome: completed | stuck | error, optional stuck_reason: needs_clarification | needs_credentials | needs_context | blocked_technical). Other products return free-text final messages and require downstream parsing.hydra reset --feedback lets the Lead actually intervene at decision points instead of being block-and-join. A stale or wrong run is one reset away.Core rules:
outcome=stuck or outcome=error.Agent launch rule:
termcanvas terminal create --prompt "...".termcanvas terminal input for task dispatch — it is not a supported automation path.Telemetry polling:
hydra watch as the main polling loop. Do not infer progress from terminal prose.termcanvas telemetry get --workflow <workflowId> --repo .termcanvas telemetry get --terminal <terminalId>termcanvas telemetry events --terminal <terminalId> --limit 20awaiting_contract means the worker has not yet published result.json; stall_candidate means the worker may be hung. Trust derived_status and task_status over terminal prose.The Lead is the decider — Lead reads the codebase, makes the strategic calls, and dispatches workers for the steps that genuinely need a fresh agent process. There is no "researcher" role: the Lead does the research
test cases passed, vs 4 / 14 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.