agentsclimarketplace

Orch state

Skill zig999/siegard-code/dist/.claude/skills/orch-state

Reduce the event log to derived state. Provides reduce.py (full OrchState JSON), summary.py (compact summary), current_phase.py (active phase), and detect_mode.py (run mode). Single source of derived state for orchestrators (P1/P2 — log is the truth, state is derived). Not user-invocable — orchestrators run the scripts directly.From its SKILL.md

Install
npx -y skills add zig999/siegard-code --skill orch-state

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

2 things 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.
  • runs commandsInstructs the agent to run 3 commands, including `python3 .claude/skills/orch-state/scripts/reduce.py` and 2 more.

SKILL.md

2.3 KB, 508 tokens by cl100k_base, as published. Nobody here has run it

orch-state

State inspection skill: reduce event log to OrchState, print summaries, query current phase.

scripts/reduce.py

Replays all events and prints the full OrchState as a JSON object. Exit 0 on success, 1 on error.

python3 .claude/skills/orch-state/scripts/reduce.py

Output schema (top-level keys):

KeyTypeDescription
workflow_idstring|nullWorkflow identifier
run_statusstringactive, escalated, etc.
current_phasestring|nullActive phase name
tasksobjectMap of task_id → TaskState
phasesobjectMap of phase name → PhaseState
escalationobject|nullEscalation payload if present
circuit_breakerobject|nullCircuit breaker state if present
last_seqintLast event seq processed
last_snapshot_seqintSeq of last snapshot event

scripts/summary.py

Prints a human-readable summary to stdout. Not JSON — intended for operator inspection.

python3 .claude/skills/orch-state/scripts/summary.py

Example output:

Workflow : wf_001
Status   : active
Phase    : dev
Last seq : 42

Tasks
─────
  pending              ████████░░░░░░░░░░░░    4 / 10
  completed            ██████████████░░░░░░    7 / 10

Tasks by phase
──────────────
  dev                  pending=4, completed=6

Phases
──────
  ▶ 1. dev                  active
    2. qa                   pending

scripts/current_phase.py

Prints the current phase and its status as JSON. Exit 0 always (errors also JSON).

python3 .claude/skills/orch-state/scripts/current_phase.py

Output when a phase is active:

{"current_phase": "dev", "status": "active", "order": 1}

Output when no phase has been entered:

{"current_phase": null, "status": null}

What ships with it: 4 files

10.0 KB alongside SKILL.md, 4 of them executable

scripts/

Keep looking

Skills are one crate of 325,949. 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.