CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

agentic-workflows

Route gh-aw workflow design/create/debug/upgrade requests to the right prompts.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.github/skills/agentic-workflows/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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.

A well-designed dispatcher body: the routing table maps task types to specific files, the workflow sequence is explicit, and a clarifying-question fallback plus overlay precedence handle edge cases. The main costs are a redundant 76-entry flat file list that inflates token usage, and an unspecified retrieval mechanism for the upstream files it says are not available locally.

Suggestions

Trim or group the flat 76-file list under "Read only the files you need" — the curated routing table already covers the actionable entries; the rest could be collapsed into category groups (e.g., syntax, safe-outputs, token-optimization) or a single reference, cutting substantial duplicated tokens.

Specify how to load files "from `github/gh-aw` (they are not available locally)" — e.g., the exact fetch command or URL pattern — so the routing step is fully executable.

Add a brief checkpoint after loading a prompt (e.g., confirm the loaded file matches the identified task type before following it) to strengthen the dispatch workflow's validation.

DimensionReasoningScore

Conciseness

The routing table, overlay rules, and OTEL carve-out are lean, but the flat 76-entry file list under "Read only the files you need" largely duplicates the curated routing bullets below it. Mostly efficient but could be tightened — anchor 3; not 2 since there is no padding that explains concepts Claude already knows.

3 / 5

Actionability

Concrete task-to-file mappings ("Design workflows from scratch via interview: `.github/aw/designer.md`"), explicit overlay loading via `@.github/aw/instructions.md`, and precedence rules give mostly executable guidance. Not 5 because the instruction to load files "from `github/gh-aw` (they are not available locally)" gives no mechanism for actually fetching them.

4 / 5

Workflow Clarity

The sequence is explicit — "identify the task type, load the matching workflow prompt/skill file, and follow it directly" — with an error-recovery checkpoint ("ask a clarifying question if the correct prompt is unclear") and overlay precedence rules. Not 5 because there is no checkpoint confirming the loaded file actually matches the identified task; this is a dispatch operation, so the destructive/batch validation cap does not apply.

4 / 5

Progressive Disclosure

All referenced files are one level deep and explicitly pathed, with no local bundle files (references/, scripts/, assets/ are absent). However, the undifferentiated 76-file flat list — which duplicates the routing table's entries — is a minor organization gap versus anchor 5's "content appropriately split; easy navigation".

4 / 5

Total

15

/

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.

A concise, router-style description that clearly states what it does and which request types trigger it, with good natural keywords around design/create/debug/upgrade. Its main weaknesses are the absent spelled-out product name (relying on the "gh-aw" abbreviation) and the omission of several task types the body actually routes (update, optimize, shared components).

DimensionReasoningScore

Specificity

"Route gh-aw workflow design/create/debug/upgrade requests to the right prompts" names the domain and several concrete request types (design, create, debug, upgrade) with a clear action (route to prompts). It is not comprehensive — the body shows additional task types (update, optimize, shared components) the description omits, matching anchor 4 rather than 5.

4 / 5

Completeness

The 'what' is clear (route requests to the right prompts) and explicit trigger guidance ("design/create/debug/upgrade requests") is present, satisfying the 'Use when' equivalence. Not 5 because the trigger guidance is embedded in the what-clause rather than stated as an explicit separate 'use when' phrasing; not 3 because the when is explicitly stated, not merely implied.

4 / 5

Trigger Term Quality

Keywords like "workflow", "design", "create", "debug", and "upgrade" match what users would naturally say. However, the spelled-out product name "GitHub Agentic Workflows" and common variations like "update" or "fix" are missing, and "gh-aw" alone is jargon — a few natural terms missing, so anchor 4 rather than 5.

4 / 5

Distinctiveness Conflict Risk

"gh-aw workflow" carves a clear niche, but the bare word "workflow" leaves minor overlap risk with generic GitHub Actions/CI workflow skills, since the description never spells out "GitHub Agentic Workflows". Mostly distinct with minor overlap — anchor 4.

4 / 5

Total

16

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

github/awesome-copilot

Path

.github/skills/agentic-workflows/SKILL.md

Reviewed

Table of Contents

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.