adding-models

Guide for adding new LLM models to Letta Code. Use when the user wants to add support for a new model, needs to know valid model handles, or wants to update the model configuration. Covers models.json configuration, CI test matrix, and handle validation.

Quality

85%

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

Quality

Content

82%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 concise, highly actionable skill body built around a clear numbered workflow with executable code at every step. Its main gaps are minor: a duplicated provider-prefix list and the absence of an explicit validation/retry loop.

Suggestions

Remove the duplicated provider-prefix list — it appears verbatim under Step 1 and again under Step 2; keep it in one place and cross-reference.

Add an explicit validation/retry loop after Step 3 (e.g., 'If the headless test fails with a handle error, re-run the curl query, correct the handle, and re-test') to lift workflow clarity.

Consider noting where the CI matrix 'around line 122' anchor may drift over time, or replace the line number with a structural anchor to avoid time-sensitive staleness.

Dimension	Reasoning	Score
Conciseness	Lean and mostly efficient with no padding of concepts Claude already knows; the only bloat is the provider-prefix list being repeated verbatim in both Step 1 and Step 2, which keeps it just below a 5.	4 / 5
Actionability	Fully executable guidance throughout — copy-paste-ready curl/jq queries, a concrete models.json entry with a field reference, a bun headless test command with an example, and a precise CI matrix line reference covering the common cases.	5 / 5
Workflow Clarity	Four clearly sequenced steps with a verification checkpoint in Step 3 (run headless to confirm the model works) and an explicit 'Optional' label on the CI step; not a 5 because there is no explicit validate-then-fix feedback loop, though the operation is config-edit rather than destructive/batch so the cap does not apply.	4 / 5
Progressive Disclosure	Well-organized single-file skill with clear sections (Quick Reference, Workflow, Toolset Detection, Common Issues) and inline pointers to real repo files (src/models.json, .github/workflows/ci.yml, src/tools/manager.ts); no external bundle files exist to signal, and no nested references, so structure is solid but the simple single-file form stops just short of the 5 anchor.	4 / 5
	Total	17 / 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.

A strong, well-formed description that pairs a concrete statement of capability with an explicit multi-trigger 'Use when' clause. It is third person and free of fluff, scoring high across all dimensions.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions ('adding new LLM models', 'add support for a new model', 'needs to know valid model handles', 'update the model configuration') plus three coverage areas (models.json, CI test matrix, handle validation); not a full 5 because the actions are stated as triggers rather than an exhaustive capability list.	4 / 5
Completeness	Clearly answers 'what' ('Guide for adding new LLM models to Letta Code. Covers models.json configuration, CI test matrix, and handle validation.') and explicitly answers 'when' via a concrete 'Use when the user wants to...' clause listing three triggers.	5 / 5
Trigger Term Quality	Good natural-term coverage with phrases users would say ('add support for a new model', 'valid model handles', 'update the model configuration'); falls short of 5 because it omits common synonyms and concrete file-extension-style identifiers.	4 / 5
Distinctiveness Conflict Risk	Clear narrow niche (adding LLM models to Letta Code) with distinct, specific triggers that are unlikely to collide with other skills; minimal conflict risk.	5 / 5
	Total	18 / 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: letta-ai/letta-code
Path: .skills/adding-models/SKILL.md
Commit: 051b47f

Reviewed: 2 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.