manage-run

Use when: User wants to INSPECT, MONITOR, START, APPROVE, or CONTROL Archon workflow RUNS in the current project — driven through the `archon` CLI over bash. Triggers (inspect): "what's running", "list runs", "show recent runs", "run status", "did the review pass", "check run <id>", "show me run <id>", "what happened in that run". Triggers (control): "approve the plan", "approve run <id>", "reject that run", "cancel that run", "abandon run <id>", "resume run <id>", "continue that run". Triggers (start): "start <workflow> in the background", "kick off <workflow> detached". Capability: Drives `archon workflow runs/get/status/run --detach/approve/reject/cancel/abandon/resume` with machine-readable `--json` output, scoped to the current project by cwd. NOT for: Authoring workflows/commands, or Archon setup/config — use the broader `archon` skill.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-structured, highly actionable CLI reference with clear sequenced workflows and a clean one-level reference split. Its main weakness is the Console UI note, which spends tokens on backend implementation details irrelevant to driving runs.

Suggestions

Trim the Console UI note: keep the user-facing fact (detached runs appear and update live in the web console, no refresh needed) and drop the server-side poller / PostgreSQL NOTIFY / SQLite interval internals.

Add a one-line validation/confirmation note for the irreversible cancel and abandon verbs (e.g., confirm the run-id via `get` before cancelling) so destructive operations have an explicit checkpoint.

Consider moving the verbose approve/resume two-step rationale into references/commands.md and keeping only the canonical command sequence inline to tighten the body further.

Dimension	Reasoning	Score
Conciseness	The body is mostly tight (a verbs table plus focused bash patterns), but the Console UI note digresses into server-side poller internals (PostgreSQL NOTIFY triggers, SQLite poller intervals) that Claude does not need to manage runs — more than a minor amount of over-explanation.	3 / 5
Actionability	Fully executable copy-paste bash throughout — the verbs table gives exact commands with flags and the patterns show real invocations (archon workflow runs --json, approve <run-id> "ship it" --json, run ... --detach) covering the common cases.	5 / 5
Workflow Clarity	Multi-step flows are clearly sequenced with polling checkpoints (poll until completed/failed/cancelled) and an error-recovery hint (check child log path if a run never appears), but the irreversible cancel/abandon verbs carry no confirmation or validation guidance — a minor validation gap.	4 / 5
Progressive Disclosure	Overview and key patterns live inline while the full flag list and JSON shapes are deferred to a single one-level-deep reference (references/commands.md, verified present), clearly signaled at the end of the body.	5 / 5
	Total	17 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: it pairs concrete capability statements with comprehensive natural-language triggers, an explicit Use-when clause, and a clear NOT-for boundary distinguishing it from the broader archon skill. Third-person voice is used correctly throughout.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions (INSPECT, MONITOR, START, APPROVE, CONTROL) and the exact CLI verbs runs/get/status/run --detach/approve/reject/cancel/abandon/resume with --json output — comprehensive coverage matching the anchor 5 example.	5 / 5
Completeness	Explicitly answers both what ("Capability: Drives archon workflow...") and when ("Use when: User wants to INSPECT, MONITOR, START, APPROVE, or CONTROL...") with concrete trigger phrases and a NOT-for boundary.	5 / 5
Trigger Term Quality	Extensive natural phrases users would actually say ("what's running", "list runs", "did the review pass", "approve the plan", "reject that run", "resume run <id>", "kick off <workflow> detached") with synonyms across approve/reject/cancel/abandon/resume/continue.	5 / 5
Distinctiveness Conflict Risk	Clear niche (managing Archon workflow runs via the archon CLI) with explicit boundary guidance ("NOT for: Authoring... use the broader archon skill") minimizing overlap with the sibling skill.	5 / 5
	Total	20 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: coleam00/Archon
Path: .claude/skills/manage-run/SKILL.md
Commit: d1ada87

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.