skcoord

Sovereign multi-agent coordination + ITIL service management, extracted from skcapstone (CR-4.1) as the standalone coordination core. A pure Python library: one runtime dependency (pydantic), no CLI, no daemon, no network surface.

Maturity-tier: T0, N/A (no key material, no crypto surface). Docs: SOP.md · SECURITY.md · CONTRIBUTING.md · CHANGELOG.md

What's here

coordination: the conflict-free task board (Board, Task, each agent writes only its own files under ~/.skcapstone/coordination/). Task files carry no status field; status is derived.
card: the read-only unified kanban Card projection over tasks + ITIL, plus the per-writer overlay log coordination/card_events/<host>.jsonl.
card_store: the event-sourced Card store (~/.skcapstone/cards/<id>/core.json written once, plus append-only per-writer event logs folded on read). SKCOORD_CARD_STORE is its kill switch, not an opt-in: post Phase-4e the store is default-ON and is disabled only by 0, off, false, or no (dual means write both, read legacy). See SOP.md section 6.
lifecycle: a read-only audit and explicit repair for drift between the event-sourced kanban lifecycle and mutable agent claim projections. Repairs refuse active ownership conflicts and append per-writer JSONL receipts.
itil: incident / problem / change / CAB / KEDB service management.
cmdb: event-sourced configuration items + relationships.
cmdb_scheduler: fail-closed scheduled-reconcile configuration, overlap lease, drift escalation, and evidence-retention helpers.
agent_card: the shareable sovereign identity vCard for the mesh.
atomic_io: atomic file-write helper shared by the above.

Dependency direction

Import-time dependencies flow one way: skcapstone depends on skcoord. The few reverse edges into skcapstone internals (skjoule, active_agent_name, gtd_tools, pubsub, activity) are runtime-lazy inside the methods that use them, so there is no import-time cycle. In skcapstone, skcapstone.coordination / .card / .card_store / .itil / .cmdb / .agent_card / .atomic_io remain as re-export shims, so every existing importer keeps working byte-identically.

That also means every skcapstone process runs skcoord code: the daemon, the per-agent units, the dashboard, and skoperator all reach it through those shims.

Install

~/.skenv/bin/pip install -e .

The version is derived from the git tag by setuptools-scm, so a shallow clone without tags builds a dev version. Use fetch-depth: 0 and fetch-tags: true.

Test

~/.skenv/bin/python -m pytest tests/ -q
~/.skenv/bin/python -m ruff check src/ tests/

License

GPL-3.0-or-later. See LICENSE.

Release files for skcoord 0.1.33

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for skcoord 0.1.33
File	Size	Uploaded
skcoord-0.1.33.tar.gz	251.8 kB	Aug 23, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for skcoord 0.1.33
File	Interpreter	ABI	Platform	Reset
skcoord-0.1.33-py3-none-any.whl 157.3 kB Aug 23, 2026	Python 3	none	any	Details

Total release size: 409.1 kB

Release files / skcoord-0.1.33.tar.gz

Download URL	skcoord-0.1.33.tar.gz
Size	251.8 kB
Tags	Source
SHA-256 checksum How to use checksums	`fe1938214c80bac520e41d69899638e2390a793965ec0ee6105dce8b9c185d8e`
BLAKE2b-256 checksum How to use checksums	`62cd9c3d65acf4e407162af4a17ff690c1c5ab915ed482828c3527fa1762944e`
Upload date	Aug 23, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/smilinTux/skcoord
Publishing commit: github.com/smilinTux/skcoord/tree/082a5077920f7126957b28f624cb8bc75e1527ea

Workflow

Publishing configuration: github.com/smilinTux/skcoord/blob/082a5077920f7126957b28f624cb8bc75e1527ea/.github/workflows/publish.yml
Publishing logs: github.com/smilinTux/skcoord/actions/runs/32661616166/attempts/1

Artifact

Subject: skcoord-0.1.33.tar.gz
SHA-256 checksum: fe1938214c80bac520e41d69899638e2390a793965ec0ee6105dce8b9c185d8e Verifying attestations

Release files / skcoord-0.1.33-py3-none-any.whl

Download URL	skcoord-0.1.33-py3-none-any.whl
Size	157.3 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`fded24d83c18758781b9686cdcdea9b42221eca41da68355e062f4c32d35d8ac`
BLAKE2b-256 checksum How to use checksums	`94c57335b4f316566401b4541bce133e9a053a36ac584e5044eac4190b4f3783`
Upload date	Aug 23, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/smilinTux/skcoord
Publishing commit: github.com/smilinTux/skcoord/tree/082a5077920f7126957b28f624cb8bc75e1527ea

Workflow

Publishing configuration: github.com/smilinTux/skcoord/blob/082a5077920f7126957b28f624cb8bc75e1527ea/.github/workflows/publish.yml
Publishing logs: github.com/smilinTux/skcoord/actions/runs/32661616166/attempts/1

Artifact

Subject: skcoord-0.1.33-py3-none-any.whl
SHA-256 checksum: fded24d83c18758781b9686cdcdea9b42221eca41da68355e062f4c32d35d8ac Verifying attestations