Tessl • The package manager for agent skills

hosted-agents

This skill should be used when designing hosted or background agent infrastructure: sandboxed execution, remote coding environments, warm pools, session persistence, multiplayer collaboration, self-spawning agents, or Modal-style sandboxes.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/hosted-agents/SKILL.md

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-structured, actionable design skill with clear sequencing and appropriate external references. The main deductions are repetition of key points across multiple sections (plus a dated metadata footer) that hurt token efficiency, and an unreferenced bundled script.

Suggestions

Consolidate repeated guidance: predictive warm-up and user-attributed commits each appear in 3-4 sections; state each once in Detailed Topics and let Guidelines/Gotchas reference it in one line.

Add a pointer to scripts/sandbox_manager.py in the body (e.g., in Sandbox Infrastructure or References) with a 'Read when' condition, so both bundle files are navigable.

Remove or trim the dated 'Skill Metadata' section (creation date, version numbers) — time-sensitive details that add tokens without aiding execution.

Dimension	Reasoning	Score
Conciseness	Domain-accurate with useful specifics (30-minute image cadence, 4-hour timeout default), but several points repeat across sections — predictive warm-up appears in Warm Pool Strategy, Speed Optimizations, Guidelines, and Gotchas, and user-attributed commits appears three times — and the dated 'Skill Metadata' section (Created 2026-01-12, Version 1.2.0) is time-sensitive information outside any 'old patterns' section. Mostly efficient but could be tightened — anchor 3, not 4.	3 / 5
Actionability	Concrete, specific guidance throughout: the numbered sandbox-to-API flow, the sequenced session lifecycle example, explicit defaults, and a checklist of decisions to make. Per the code-vs-instruction note, absence of inline code is not penalized since the guidance is actionable and a bundled executable script exists. Minor gaps (no inline pointer to the script) keep it below 5.	4 / 5
Workflow Clarity	Multi-step processes are clearly sequenced (warm sandbox allocation -> delta sync -> reads-unblocked/writes-gated -> snapshot -> push -> PR) with most checkpoints present ('check token validity before sensitive operations', 'snapshot before termination', 'alert if builds fail silently'). Not 5 because there is no explicit validate -> fix -> retry feedback loop.	4 / 5
Progressive Disclosure	Well-organized overview with a one-level-deep, clearly signaled reference to references/infrastructure-patterns.md including a 'Read when' condition. Scored against the actual bundle: scripts/sandbox_manager.py exists but is never mentioned in the body, a minor organization gap — anchor 4, not 5.	4 / 5
	Total	15 / 20 Passed

Description

66%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 well-targeted description with strong trigger coverage for a niche domain. Its main weakness is that it is purely a 'when' clause: the enumeration of infrastructure topics implies but never states what the skill actually provides, capping completeness.

Suggestions

Lead with an explicit 'what' clause stating what the skill provides (e.g., 'Provides design guidance and implementation patterns for hosted agent infrastructure'), keeping the existing trigger enumeration as the 'when'.

Clarify scope boundaries in the description itself (e.g., 'infrastructure and runtime concerns only, not agent loop or topology design') to reduce overlap with adjacent skills like multi-agent-patterns.

Add one or two more common trigger variations users might naturally say, such as 'cloud sandboxes' or 'agent runtime infrastructure'.

Dimension	Reasoning	Score
Specificity	Enumerates several concrete sub-domains ('sandboxed execution, remote coding environments, warm pools, session persistence, multiplayer collaboration, self-spawning agents, or Modal-style sandboxes') but never states what the skill itself does. Not 5 because there is no list of concrete actions performed; not 3 because far more than 1-2 specific items are named.	4 / 5
Completeness	The 'when' is explicit and rich ('This skill should be used when designing hosted or background agent infrastructure: ...'), but the 'what' is entirely implicit — the description never states what the skill provides or does. Falls between anchor 2 (only 'when' present) and anchor 3; the explicit domain enumeration weakly implies the what, so 3 fits best.	3 / 5
Trigger Term Quality	Natural terms a user designing this infrastructure would say ('hosted', 'background agent', 'warm pools', 'Modal-style sandboxes') with synonyms ('remote coding environments'). Not 5 because a few common variations like 'cloud sandbox' or 'agent runtime' are missing.	4 / 5
Distinctiveness Conflict Risk	'Hosted or background agent infrastructure' with 'Modal-style sandboxes' is a clear niche with distinct triggers, but 'self-spawning agents' and 'multiplayer collaboration' overlap adjacent skills like multi-agent-patterns. Mostly distinct with minor overlap risk — anchor 4, not 5.	4 / 5
	Total	15 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: muratcankoylan/Agent-Skills-for-Context-Engineering
Path: skills/hosted-agents/SKILL.md
Commit: 58b55a8

Reviewed: 11 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.