CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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

76%

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 ./claude/plugins/agentlas-core-engine-meta-agent/runtime-support/skills/agentlas-packaging/SKILL.md

The canonical home for this skill is agentlas-packaging in agentlas-ai/Agentlas-OS

SKILL.md
Quality
Evals
Security

Quality

Content

63%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 an efficiently written, well-sequenced procedure with real validation checkpoints, but its guidance is more catalog-than-executable: only one runnable command, missing detail on how to author the contracts it mandates, and references to bundle files that are not actually present. Step 5's inlined ontology spec should live in a reference file.

Suggestions

Include the referenced bundle files (contracts/builder-interview-research-gate.md, scripts/verify-team-package.sh) in the skill, or remove/replace the references — pointing at nonexistent files breaks the procedure at steps 3 and 8.

Move step 5's workforce résumé ontology constraints (roles/communities/skills/knowledge ranges, the ontology_v1.json reference) into a dedicated reference file and keep only a summary in SKILL.md, improving both actionability context and progressive disclosure.

Add concrete authoring detail for the key contracts — an example agent-card.json workforce block, the shape of a runtime adapter, and how to run card lint — so steps 4-6 are executable rather than descriptive.

DimensionReasoningScore

Conciseness

The body is lean and imperative — numbered steps, file manifests, no explanations of concepts Claude already knows. Only minor trimming is needed: the rationale sentence in step 3 ("Packaging must not turn a shallow prompt into a polished but weak package") and step 9, which partially duplicates step 8's verification. Not a 5 due to that redundancy; not a 3 since the padding is genuinely minor.

4 / 5

Actionability

Concrete elements exist — one executable command ("scripts/verify-team-package.sh <package-root>") and a specific manifest of contract file paths — but most steps lack execution specifics: nothing shows the shape of agent-card.json, what a runtime adapter or global command concretely looks like, or how to invoke "card lint". This matches the 3 anchor (some concrete guidance, missing key details) rather than the mostly-executable 4.

3 / 5

Workflow Clarity

A clear 9-step sequence (inspect, classify, gate, add contracts, résumé block, global command, sanitize, verify, release checks) with an explicit validation checkpoint and feedback loop in step 8 ("If it fails, correct the package shape before any final handoff"). Validation is present so the batch/destructive cap at 3 does not apply, but the 5 anchor is missed because step 3's research gate has no exit criteria and step 9's public-safety checks are unspecified.

4 / 5

Progressive Disclosure

The body references external files ("contracts/builder-interview-research-gate.md", "scripts/verify-team-package.sh") that do not exist in the bundle — the referenced directories are absent entirely — and step 5 inlines ~16 lines of workforce-ontology contract detail that belongs in a reference file. Structure exists (Procedure, Output sections) with clearly signaled paths, matching the 3 anchor rather than the well-split 4.

3 / 5

Total

14

/

20

Passed

Description

78%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 strong description: it states concrete actions, an explicit trigger clause, and a well-branded niche unlikely to collide with other skills. Its main limitation is that the "when" clause restates the actions rather than giving distinct trigger contexts, and a few natural synonyms are absent.

DimensionReasoningScore

Specificity

Names three concrete actions ("converting, repairing, or packaging") applied to a defined target ("existing local or external agent/team") across five enumerated destination contexts. Minor gaps remain — the description never hints at what packaging entails (contract authoring, verification) — so it falls short of the comprehensive 5 anchor but sits clearly above the 1-2 action coverage of a 3.

4 / 5

Completeness

Both parts are present: the "what" (convert/repair/package an agent into Agentlas architecture) and an explicit "Use when..." clause. The "when" is explicit but coextensive with the action verbs themselves rather than naming distinct user-side trigger contexts (e.g., "when the user mentions Agentlas"), matching the 4 anchor rather than the 5.

4 / 5

Trigger Term Quality

Contains natural terms users would plausibly say ("converting", "packaging", "repairing", "agent/team", "Codex plugin", "Claude adapter", "open-source release") with good coverage of delivery contexts. A few natural synonyms like "port", "bundle", or "publish" are missing, matching the 4 anchor rather than the comprehensive 5.

4 / 5

Distinctiveness Conflict Risk

The "Agentlas" branding plus specific delivery targets (Agentlas import, Codex plugin use, Claude adapter use) carve out a clear niche with distinct triggers and minimal conflict risk; only "open-source release" mildly overlaps with generic release-packaging skills.

5 / 5

Total

17

/

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

claude/plugins/agentlas-core-engine-meta-agent/runtime-support/skills/agentlas-packaging/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.