CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

hephaestus-build

Use when the user types /prompts:hep-build or /agentlas-build, mentions @Hephaestus for build work, asks to create a single Agentlas agent, create a multi-agent team, or package an existing local/external agent into Agentlas architecture.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

70%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 genuinely actionable procedure with best-in-class validation and failure-recovery gating, and it avoids teaching concepts Claude already knows. Its weaknesses are token efficiency and organization: redundant steps and inline policy detail make it long, and everything lives in one file with no reference bundle or clear navigation layer.

Suggestions

Split steps 13-14's storage and upload policy detail (visibility options, secret-review file rules, failure/retry handling) into a reference file such as references/upload-policy.md, keeping only the decision question and gating rule inline.

Remove the redundancy between step 3's mode classifier and step 6's "Pick one" re-listing by merging them into a single classify-and-commit step.

Move the interview question phrasing and research-gate bullet list from step 4 into a reference file (or defer fully to the engine contract), and name the specific support skills to load in step 7.

DimensionReasoningScore

Conciseness

The body is operational rather than padded, but at ~160 lines it has real redundancy — step 6 re-lists the mode choice already classified in step 3, step 2's session handling overlaps step 4's, and steps 13-14 inline full storage/upload policy detail (e.g., "Mobile can use the package only after a paired Desktop restores/installs it; Cloud is not a hosted LLM runtime") that could be tightened or moved to references.

3 / 5

Actionability

Provides copy-paste-ready bash for engine/runner resolution and concrete runner commands ("$RUNNER" contract resolve-target/scaffold/complete/verify, hep-upload with --visibility flags) plus exact file paths and output fields, but has minor gaps: "--mode <single|team|package>" placeholders, step 7 ("Load matching support skills") names no skills, and "/hep-upload Step 3" is referenced but not included.

4 / 5

Workflow Clarity

The 14-step sequence is explicit with thorough validation gates and error-recovery loops: status-ok receipt required ("Nonzero or error receipts are blockers"), verify-team-package.sh failure handling ("If it fails, do not report completion; collapse the output... or add the required orchestrator/HQ"), contract verify gating ("A non-zero exit or non-empty blocker list means blocked, never completed"), and upload failure handling ("Keep the local package on every auth/offline/CAS/quota/scan failure and report an exact retry command").

5 / 5

Progressive Disclosure

The skill is a single monolithic file with no bundle; it wisely delegates detailed contract content to engine files ($ENGINE/contracts/builder-interview-research-gate.md, mode contracts), but those references are buried mid-procedure rather than clearly signaled, and inline policy blocks (interview question lists, storage policy, upload rules) would be better split into reference files.

3 / 5

Total

15

/

20

Passed

Description

87%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 strong: it explicitly pairs a "Use when..." trigger list with three concrete capabilities, all anchored to a distinctive branded niche. The only weakness is that it undersells the skill's full scope (repair, verification, upload/publish), which slightly caps specificity and trigger coverage.

DimensionReasoningScore

Specificity

Lists three concrete actions ("create a single Agentlas agent, create a multi-agent team, or package an existing local/external agent into Agentlas architecture") but omits capabilities the body covers such as repair, verification, and Cloud/Hub publishing, so coverage has minor gaps rather than being comprehensive.

4 / 5

Completeness

Explicitly answers both questions: the "Use when..." clause gives concrete trigger conditions and the action list states what the skill does, matching the top anchor's structure of explicit what-and-when with concrete trigger phrases.

5 / 5

Trigger Term Quality

Includes concrete trigger phrases users would naturally say or type ("/prompts:hep-build", "@Hephaestus for build work", "create a multi-agent team", "package an existing local/external agent"), but misses natural variations for body-covered capabilities such as "repair", "publish", and "upload".

4 / 5

Distinctiveness Conflict Risk

Every trigger is anchored to the niche Agentlas/Hephaestus domain with unique slash commands (/prompts:hep-build, /agentlas-build) and an @mention, giving a clear niche with minimal conflict risk against other skills.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

agentlas-ai/Agentlas-OS

Path

codex/plugins/agentlas-core-engine-meta-agent/skills/hephaestus-build/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.