AF goal-runner
Autonomous multi-task orchestration loop over a roadmap file (ROADMAP.md, TODO.md, or any checkbox task list) — pairs with the built-in /goal Stop hook to work tasks ONE at a time until the file is done. The main agent orchestrates only; every task is delegated to sub-agents for coding, code review, and regression testing, with observed-green gates before the checkbox is ticked. No commits or pushes unless the goal text explicitly authorizes them — when it does, one commit per verified task, never red. Blocked tasks are annotated and skipped, never faked. Use this skill whenever the user says "work on the roadmap tasks til completion", "work the roadmap until completion", "work on tasks until completion using sub agents", "launch sub agents to work the tasks", "/goal-runner", or sets a /goal naming a roadmap/task file — even if they don't name the skill. Not for a single defined task (use autopilot or task-executor) and not for interval-scheduled reruns (use loop).
Autonomous multi-task orchestration loop over a roadmap file (ROADMAP.md, TODO.md, or any checkbox task list) — pairs with the built-in /goal Stop hook to…
As a process F 58/100 · Will not run — References files that are not bundled: scripts/scan.sh
How to improve
- The text references files that are not there: add them or drop the references.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
missing-refreference to a missing file: scripts/scan.sh
Process rating: all ten parameters 58/100
- 0Tools and files. 1 referenced file(s) missing: scripts/scan.sh
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 56 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 70Execution cost. Instruction body is 4710 tokens
- 100Steps. 59 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 2 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (14 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +3Description length 979: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -5TODO / placeholder text left in the skill
- -217 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 5 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 14 headings
- +3Step-by-step instructions: 59 items
- +4Has examples (0 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.