iii-agentic-backend
Creates and orchestrates multi-agent pipelines on the iii engine. Use when building AI agent collaboration, agent orchestration, research/review/synthesis chains, or any system where specialized agents hand off work through queues and shared state.
76
70%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/iii-agentic-backend/SKILL.mdQuality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid description that clearly identifies its niche (multi-agent pipelines on the iii engine) and provides explicit trigger guidance via a 'Use when' clause with good keyword coverage. Its main weakness is that the 'what' portion could be more specific about the concrete actions performed beyond 'creates and orchestrates'—listing specific capabilities like defining agent roles, configuring handoff queues, or managing shared state would strengthen it.
Suggestions
Expand the capability list with more specific concrete actions, e.g., 'Defines agent roles, configures handoff queues, manages shared state, and builds research/review/synthesis chains on the iii engine.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain ('multi-agent pipelines on the iii engine') and mentions some actions ('creates and orchestrates'), but doesn't list multiple specific concrete actions like defining agent roles, configuring queues, setting up shared state schemas, etc. | 2 / 3 |
Completeness | Clearly answers both 'what' (creates and orchestrates multi-agent pipelines on the iii engine) and 'when' (explicit 'Use when' clause covering agent collaboration, orchestration, research/review/synthesis chains, and systems with queue-based handoffs). | 3 / 3 |
Trigger Term Quality | Good coverage of natural terms users would say: 'multi-agent', 'agent collaboration', 'agent orchestration', 'research/review/synthesis chains', 'queues', 'shared state', 'specialized agents', 'hand off work'. These are terms a user building such systems would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to the specific platform ('iii engine') and the narrow domain of multi-agent pipeline orchestration with queues and shared state. Unlikely to conflict with general coding or single-agent skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a reasonable architectural overview of multi-agent orchestration on the iii engine with clear conceptual framing and a useful primitives table. However, it lacks inline executable code examples, relying entirely on an external reference file, and the workflow description omits error handling and validation checkpoints. There is also notable redundancy between the primitives table and the common patterns section.
Suggestions
Add at least one inline, executable code snippet showing a minimal two-agent pipeline (e.g., researcher → critic) with queue handoff, rather than relying solely on the external reference file.
Add explicit error handling and validation steps to the workflow — what happens when an approval gate rejects, when an agent fails, or when state is missing/inconsistent.
Consolidate the 'iii Primitives Used' table and 'Common Patterns' section to eliminate redundancy — one or the other, not both.
Move the generic 'When to Use' and 'Boundaries' boilerplate to frontmatter or remove it, as it consumes tokens without adding actionable guidance.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient but has some redundancy — the 'Common Patterns' section largely repeats the 'iii Primitives Used' table, and the 'When to Use' and 'Boundaries' sections at the end are generic boilerplate that adds little value. The 'Adapting This Pattern' section is reasonably lean but could be tighter. | 2 / 3 |
Actionability | The skill provides API signatures and a primitives table, but no inline executable code examples. It relies entirely on an external reference file (agentic-backend.js) for the actual implementation. The 'Common Patterns' section lists API calls but without concrete, copy-paste-ready code blocks showing how they fit together. | 2 / 3 |
Workflow Clarity | The ASCII architecture diagram shows the multi-step agent pipeline sequence clearly, and approval gates are mentioned. However, there are no explicit validation checkpoints or error recovery/feedback loops described — what happens if an agent fails, if the approval gate rejects, or if state is inconsistent is not addressed. | 2 / 3 |
Progressive Disclosure | The skill references external files (agentic-backend.js, iii-config.yaml) for detailed content, which is good structure. However, no bundle files were provided to verify these references exist, and the inline content has some redundancy between sections (Primitives table vs Common Patterns) that could be better organized. The pattern boundary and 'when to use' sections feel like they belong in frontmatter rather than body content. | 2 / 3 |
Total | 8 / 12 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 — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- iii-hq/iii
- Commit
d51a06d
- 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.