Question 1

How do I update .spec-context.json after running speckit tasks?

Accepted Answer

Run the write-context.py script with --step tasks --status ready-to-implement from the repository root. The script resolves the active feature directory automatically and writes the state atomically, appending to the history array.

Question 2

How does the companion script resolve the active feature directory?

Accepted Answer

It checks in order: an explicit --feature-dir flag, the SPECIFY_FEATURE_DIRECTORY environment variable, the SPECIFY_FEATURE variable, .specify/feature.json, and finally the current git branch prefix.

Question 3

What happens if python3 is not installed when the capture hook runs?

Accepted Answer

The capture is skipped with the warning '[companion] Warning: python3 not detected; skipped .spec-context.json capture'. The host command does not fail, since the hook is state-writing only and best-effort.

Question 4

Does the capture-tasks hook create the tasks document itself?

Accepted Answer

No. The tasks document is created by the core /speckit.tasks workflow. This hook only writes the completion state into .spec-context.json so the Companion GUI re-renders with the correct step and status.

Question 5

Why does the capture script exit without writing anything?

Accepted Answer

If the active feature directory cannot be resolved from any of the fallback sources, the script prints a warning to stderr and exits 0 without writing. Pass --feature-dir explicitly to make resolution unambiguous.

speckit-companion-capture-tasks

SYSTEM DOCUMENTATION & REQUIREMENTS

Frequently Asked Questions about speckit-companion-capture-tasks

Related Skills

speckit-git-feature

swain-do

pattern-memento

eval-speckit-extension