agent-orchestration-advisor

Design multi-agent AI workflows with clear boundaries, handoffs, and monitoring. Use when a complex PM task should run as parallel specialized agents instead of one linear process.

Quality

61%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/agent-orchestration-advisor/SKILL.md

Quality

Content

48%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-sequenced interactive advisor but is heavily verbose and monolithic, re-explaining well-known concepts and inlining material that belongs in separate reference files. Actionability is moderate thanks to structured questions and examples, but it is undercut by template placeholders and abstract framing.

Suggestions

Trim or remove the concept-review sections (PM-vs-orchestration table, Teresa Torres AI-shaped problems, Marty Cagan four risks, anti-patterns) that restate knowledge Claude already has; keep only the orchestration-specific guidance.

Move the three full worked examples and the external-frameworks/tooling references into separate files under references/ and link to them one level deep, so the main body is a lean overview.

Replace the Step 8 ASCII template that restates already-collected fields with a compact generated-plan instruction, and add an explicit validation checkpoint between Step 2 (decomposition) and Step 3 (topology) so the design is confirmed before flows are drawn.

Dimension	Reasoning	Score
Conciseness	The ~765-line body extensively explains concepts Claude already knows (PM-vs-orchestration tables, Teresa Torres 'AI-shaped problems', Marty Cagan's four risks, anti-patterns) and repeats scaffolding ('Agent asks:' / 'User response: [Selection]') plus a giant ASCII template that restates already-collected fields, making it noticeably verbose with several padded sections though not entirely devoid of instructional value.	2 / 5
Actionability	Provides concrete guidance in places (named example decompositions, three flow patterns with ASCII diagrams, mission/constraints/priorities/evidence checklists, eval-type tables), but much remains abstract framing and template placeholders ([Your Task Name], [List]) with light tooling specifics for actually running an orchestration.	3 / 5
Workflow Clarity	A clearly sequenced Steps 0–8 with a defined deliverable ('Agent provides:') per step; the skill is advisory rather than destructive/batch so the destructive-cap does not apply, though explicit validation/feedback checkpoints between steps (e.g., confirm decomposition before topology design) are largely implicit.	4 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent) and the skill is a single monolithic ~765-line file with three full worked examples, an external-frameworks list, and a deep four-dimension breakdown all inlined rather than offloaded; section headers give some structure but content that should be separate remains inline.	3 / 5
	Total	12 / 20 Passed

Description

75%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 third-person, concrete, and cleanly answers both what and when, with a distinct niche. It scores 4 across all dimensions because each is solid but slightly short of the top anchor's comprehensiveness in concrete actions, trigger synonyms, and explicit multi-trigger 'when' phrasing.

Dimension	Reasoning	Score
Specificity	Names the domain ('multi-agent AI workflows') and three concrete design actions ('clear boundaries, handoffs, and monitoring'), landing just below the comprehensive 5-anchor which expects multiple distinct concrete actions across the full surface.	4 / 5
Completeness	Explicitly answers both 'what' (design multi-agent workflows with boundaries/handoffs/monitoring) and 'when' ('Use when a complex PM task should run as parallel specialized agents instead of one linear process'); the 'when' is concrete but framed narrowly rather than enumerating multiple user-side triggers, keeping it just below the 5-anchor.	4 / 5
Trigger Term Quality	Includes natural trigger phrases ('parallel specialized agents instead of one linear process') but misses common variations a PM might actually say (e.g., 'AI workflow', 'automate my research', 'break this into agents'), so it is above the 'some relevant keywords' anchor but short of full synonym coverage.	4 / 5
Distinctiveness Conflict Risk	The 'parallel specialized agents instead of one linear process' framing carves a clear niche distinct from sibling skills like context-engineering-advisor, with only minor overlap risk among closely related orchestration/advisor skills.	4 / 5
	Total	16 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (783 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 5 suspicious	Warning

	Total	13 / 16 Passed

Repository: deanpeters/Product-Manager-Skills
Path: skills/agent-orchestration-advisor/SKILL.md
Commit: 1b5a524

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.