agentsclimarketplace

Phase example

Skill zig999/siegard-code/dist/.claude/skills/phase-_example

Most AI coding tools help you write code. Siegard Code manages the entire development lifecycle — it writes specifications, plans backlogs, implements features, runs QA, and delivers tested code. All autonomously, all traceable, all through Claude Code.

Install
npx -y skills add zig999/siegard-code --skill phase-_example

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

One thing to look at

  • 9 stars9 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Scaffold template for phase-{name}-rules skills. Copy the directory to phase-{name}-rules, replace {name} placeholders, implement the exit-criteria checkers and select_worker.py routing, and declare exit-criteria.json. Reference template only — never dispatched at runtime. Not user-invocable.

SKILL.md

3.2 KB, as published. Nobody here has run it

phase-{name}-rules

Phase rules skill for {name} phase. Provides exit criteria checkers, worker routing table, and phase-specific constraints consumed by the orchestrator.

Contract

Input (from orchestrator)

The orchestrator calls this skill's scripts directly. No inter-skill communication envelope needed.

Output

Every script in scripts/ returns a JSON object to stdout and exits 0 on success or 1 on error.


Phase identity

FieldValue
phase_name{name}
order{N}
requiredtrue
worker_default{worker-agent-name}

Worker routing table

Maps task.type to worker sub-agent. Consumed by the orchestrator dispatcher (Step 5).

task.typeworker subagent_type
{type_a}{worker-a}
{type_b}{worker-b}
* (default){worker-default}

scripts/select_worker.py

Returns the worker sub-agent name for a given task type.

Usage

python3 .claude/skills/phase-{name}-rules/scripts/select_worker.py \
  --task-type <type>

Output

On success (exit 0):

{"worker": "<subagent-name>", "task_type": "<type>", "phase": "{name}"}

On error (exit 1):

{"status": "error", "reason": "unknown_task_type", "detail": "<message>"}

scripts/check_{criterion_name}.py

Evaluates one exit criterion. Replace {criterion_name} with the actual criterion identifier (e.g. all_impl_tasks_terminal, all_specs_approved, no_open_critical_findings).

One script per criterion. Each script is independent and has no side effects.

Usage

python3 .claude/skills/phase-{name}-rules/scripts/check_{criterion_name}.py

Output schema

{
  "criterion": "{criterion_name}",
  "met": true,
  "evidence": {
    "total": 0,
    "passing": 0,
    "failing": []
  }
}
FieldTypeDescription
criterionstringCriterion identifier (matches key in exit-criteria.json)
metbooltrue if criterion is satisfied
evidenceobjectSupporting data — structure varies by criterion

On error (exit 1):

{"status": "error", "reason": "<code>", "detail": "<message>"}

Error reason codes:

  • log_missing.orch/log.jsonl not found
  • internal_error — unexpected runtime error

exit-criteria.json

Declares all exit criteria for this phase. The orchestrator evaluates each criterion by calling the corresponding check_{criterion_id}.py script.

See exit-criteria.json in this directory.


references/

Phase-specific reference documents. Optional — add files here when the phase has domain-specific rules, templates, or vocabulary that workers need to consume.

FilePurpose
(empty in this template)(add as needed)

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.