agentlas-packaging

Use when converting, repairing, or packaging an existing local or external agent/team into Agentlas architecture for local install, Agentlas import, Codex plugin use, Claude adapter use, or open-source release.

Quality

73%

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 ./.agents/skills/agentlas-packaging/SKILL.md

Quality

Content

62%

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

The skill has a clear sequenced workflow with real validation checkpoints, but it reads as an instruction outline rather than an actionable guide: steps name contracts and docs to add without executable detail, the workforce block is dense inline material, and the many referenced files are not bundled or clearly navigable.

Suggestions

Provide an executable example for the core action — e.g. a concrete invocation of `scripts/verify-team-package.sh` with expected pass/fail output, and a minimal `.agentlas/agent-card.json` skeleton — so steps are copy-paste ready rather than descriptive.

Move the dense workforce ontology detail (counts, versioning, seed-snapshot caveat) into a dedicated reference file and keep SKILL.md a lean overview that links to it, improving both conciseness and progressive disclosure.

Bundle or clearly signal the referenced files (docs/*, .agentlas/*, scripts/verify-team-package.sh) as one-level-deep references with working links so navigation is concrete rather than implied paths.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and avoids re-explaining concepts Claude knows, but step 5's dense workforce block (roles/communities/skills/knowledge counts, ontology versioning, and seed-snapshot caveats) is heavy inline detail that earns its place only partially and could be tightened or moved out.	2 / 3
Actionability	It points to concrete artifacts (`docs/builder-interview-research-gate.md`, `scripts/verify-team-package.sh`) and an exact ontology path, but most steps describe rather than instruct — there is no executable code or command for steps 1-7, only file/contract names to add, so guidance is concrete but incomplete.	2 / 3
Workflow Clarity	The procedure is a clearly numbered 9-step sequence with an explicit research-gate checkpoint (step 3) and a verification gate with a fix-before-handoff feedback loop (steps 8-9), matching the explicit-validation-and-feedback-loop anchor.	3 / 3
Progressive Disclosure	It references many external paths (docs/, .agentlas/, scripts/verify-team-package.sh, agentlas_cloud/workforce/ontology_v1.json), but no bundle files (references/, scripts/, assets/) are present and none of the referenced paths are signaled as one-level-deep navigable links — references are inline path mentions with no structure confirming they exist or how to reach them.	2 / 3
	Total	9 / 12 Passed

Description

85%

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: concrete actions, an explicit "Use when" trigger, and a distinct Agentlas-specific niche. Its main weakness is trigger-term naturalness — much of the wording is product jargon rather than language a user would spontaneously say.

Suggestions

Add natural-language trigger phrasings a user would actually say (e.g. "Use when the user asks to package an agent for Claude Code or Codex, or to convert an existing agent into Agentlas format") alongside the current product-jargon list.

Soften or gloss the proprietary terms ("Agentlas architecture", "workforce") so non-expert users can recognize when this skill applies.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — "converting, repairing, or packaging" — and enumerates five distinct target destinations ("local install, Agentlas import, Codex plugin use, Claude adapter use, or open-source release"), matching the multiple-specific-actions anchor.	3 / 3
Completeness	It states both what the skill does ("converting, repairing, or packaging an existing... agent/team into Agentlas architecture") and when to use it via an explicit "Use when..." clause with enumerated trigger conditions, satisfying the both-what-and-when anchor.	3 / 3
Trigger Term Quality	It names the domain ("Agentlas architecture") and verbs a user might say ("packaging", "converting"), but the surrounding terms ("Agentlas import", "Codex plugin", "Claude adapter") are product jargon rather than natural trigger phrasings a user would spontaneously say, so coverage is partial.	2 / 3
Distinctiveness Conflict Risk	The Agentlas-specific niche and enumerated target runtimes (Codex plugin, Claude adapter) make it a clear, distinct domain unlikely to fire for unrelated packaging skills.	3 / 3
	Total	11 / 12 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
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: agentlas-ai/Agentlas-OS
Path: .agents/skills/agentlas-packaging/SKILL.md
Commit: 2eb39ad

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.