Hk local diagnose

Skill deepklarity/harness-kit/.claude/skills/hk-local-diagnose

A kit for building with AI agents and also the engineering patterns around it.

Install

npx -y skills add deepklarity/harness-kit --skill hk-local-diagnose

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Run diagnostic scripts on tasks, specs, boards, or reflections. Wraps testing_tools/ with proper working directory and output mode selection. Use this skill whenever the user wants to inspect, debug, or check the state of a task, spec, board, or reflection — even if they don't say 'inspect' explicitly. Triggers on: 'why did task X fail', 'show me spec Y', 'what's on the board', 'check task', 'inspect', 'diagnose', or /hk-local-diagnose.

SKILL.md

3.2 KB, as published. Nobody here has run it

/hk-local-diagnose — Diagnostic Script Runner

Run the right diagnostic script with the right flags, from the right directory. No more remembering paths or cd-ing around.

Usage

/hk-local-diagnose task 42
/hk-local-diagnose task 42 --brief
/hk-local-diagnose spec 15 --json --sections tasks,problems
/hk-local-diagnose board
/hk-local-diagnose board 3
/hk-local-diagnose reflection 8 --full
/hk-local-diagnose snapshot sp25 ../../tests/e2e_snapshots/smoke

Arguments

Parse $ARGUMENTS to extract:

type (required): task, spec, board, reflection, or snapshot
id (required for all except board): the numeric ID or spec prefix
flags (optional): --brief, --full, --json, --slim, --sections <list>

If no flags are provided, default to --brief — this is the token-efficient choice for LLM consumption. The user can always ask for --full if they need more.

Script mapping

Type	Script	Required args
`task`	`task_inspect.py <id>`	id
`spec`	`spec_trace.py <id>`	id
`board`	`board_overview.py [id]`	id optional
`reflection`	`reflection_inspect.py <id>`	id
`snapshot`	`snapshot_extractor.py <id> <output_dir>`	id + output_dir

Execution

All scripts run from the taskit/taskit-backend/ directory. The working directory for execution is always:

REPO_ROOT/taskit/taskit-backend/

Where REPO_ROOT is the git repository root (find it with git rev-parse --show-toplevel).

Step 1: Resolve the repo root

REPO_ROOT=$(git rev-parse --show-toplevel)

Step 2: Build and run the command

cd "$REPO_ROOT/taskit/taskit-backend" && python testing_tools/<script> <id> [flags]

Pass through any --brief, --full, --json, --slim, or --sections flags directly to the script.

Step 3: Display the output

Print the script output directly. Do not summarize or interpret — the scripts already produce well-structured output with problem detection built in.

If the script exits with a non-zero code, show the error and suggest:

Check that the ID exists: "Is task/spec/board {id} a valid ID?"
Check that the Django app is set up: "Is the taskit-backend database accessible?"

Error handling

If $ARGUMENTS is empty or missing the type, print usage help:

Usage: /hk-local-diagnose <type> <id> [flags]

Types: task, spec, board, reflection, snapshot
Flags: --brief (default), --full, --json, --slim, --sections <list>

Examples:
  /hk-local-diagnose task 42
  /hk-local-diagnose spec 15 --json --sections tasks,problems
  /hk-local-diagnose board