Skip to content
Skills Library
AI & agents · Automation

protocol-task

/protocol-task

Use when the coordinator asks a paired node to run one read-only task and the result has to come back signed and typed.

Licence
Apache-2.0
Last repository change
04.11.2026

Assessment and security

85/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content20 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
2
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 04.11.2026, without references to retired models or APIs.

Licence

Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.

What it does and when to use it

Use when the coordinator asks a paired node to run one read-only task and the result has to come back signed and typed.

Original author description

Use when the coordinator asks a paired node to run one read-only task and the result has to come back signed and typed.

How it works

  1. created, sent, accepted, running, done, failed, rejected, timedout, cancelled
Original SKILL.md headings · 10 lines, 182 words
SKILL.md preview
---
name: protocol-task
description: Use when the coordinator asks a paired node to run one read-only task and the result has to come back signed and typed.
---

# A read-only task from the coordinator to a node

The owner (or code, for the node's self-report on connect) asks the coordinator to run a task on ONE connected node. The coordinator sends the task inside the node's session with a capability token it signed: the task, the node, the session, the tool allowed, single use, and a time to live the node measures on its own clock. The node checks the token and its own read-only allowlist, then runs the tool and returns a signed, typed result with evidence.

Any task that is not on the allowlist (a write, a shell command, anything else) is rejected with a typed reason. A node that is not connected gets a failed task at once; tasks are not queued, because a queued task would run in a session its token was not made for. At most a small number of tasks run on a node at once.

## created, sent, accepted, running, done, failed, rejected, timed_out, cancelled
The states are enforced in code (`envrec_net.protocol`); a move without its evidence is refused and logged.
Files in this package (2)
  • SKILL.md 1 KB
  • skill.meta.json 5 KB

Versions

The database has no other skills named protocol-task.

Database snapshot: 13.11.2026 · package version f162b513da05 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 85protocol-analyze
    Use when the coordinator has to find out what a paired machine is, by picking read-only probes one at a time; the analysis ends with finish_analysis and is stored for the setup-options protocol.
  • 100coordinator
    Act as a coordinator instead of doing the work yourself - split a task, run the pieces as background agents with a model and effort suited to each, collect their results, get independent review, send fixes back to the author, merge, verify the results yourself, and show the user…
  • 90pi-coordinator
    Run as coordinator that plans a job, hands each coding task to a sub-agent worker (pi Agent tool / omp task agent), verifies every result itself, and lands passing work on local main as one squash commit per task. Use this whenever the user says "use pi", "spawn pi workers", "be…
  • 90swarm-coordinator
    How to hand a goal to a p2p-swarm and get the result back - writing a job the swarm can run with on its own, following progress, and stepping in only when needed. Use when the user wants to post a job, farm work out to other Claude sessions, or check on a job.
  • 100pi-coordinator
    Run as coordinator that plans a job, hands each coding task to a separate `pi` worker over its RPC mode, watches the workers with a Monitor, verifies every result itself, and lands passing work on local main as one squash commit per task. Use this whenever the user says "use…
  • 80Coordinator
    Run as the human's coordinator: the one agent that plans, dispatches Orca workers, judges their reports, keeps the board and the hand-over state, and asks the human only for sensitive decisions. Load when the user asks you to coordinate, dispatch or supervise, or when a…

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.