Tessl • The management layer for AI agent skills

SkillPublic

prompt-optimization

Applies prompt repetition to improve accuracy for non-reasoning LLMs. Use when a Haiku or other non-reasoning agent misses details in a long prompt.

Quality

80%

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 ./examples/agent-skills/prompt-optimization/SKILL.md

Quality

Content

71%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 content is well-structured, reasonably lean, and concrete for an automatic skill, with a clear mechanism example and usable configuration. Its main defects are a dangling reference to a nonexistent references/prompt-repetition.md file and mild redundancy between the activation and agent-instruction sections.

Suggestions

Fix or remove the broken reference to references/prompt-repetition.md — either create the file or delete the References section, since a dangling pointer currently breaks the skill's only external navigation.

Merge the 'Agent Instructions' section into 'When to Activate' to eliminate near-duplicate content and tighten token usage.

Add a one-line way to confirm the optimization is applied (e.g., how to check LOKI_PROMPT_REPETITION status or the metrics files) so the skill includes a verification checkpoint.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient — short sections, a compact BEFORE/AFTER example, and tables — but the 'Agent Instructions' section largely duplicates 'When to Activate' and the Metrics section adds little actionable value, leaving minor trimmable content.	4 / 5
Actionability	The BEFORE/AFTER prompt diff and concrete environment variables (e.g., LOKI_PROMPT_REPETITION_COUNT=2) give mostly executable guidance, but the skill omits how to verify repetition is active or handle edge cases, and offers no actionable step for the agent beyond 'No action needed from you'.	4 / 5
Workflow Clarity	As a single-mechanism skill the agent action is unambiguous ('Your prompt will be automatically repeated 2x... No action needed from you'), but the body never clarifies who performs the repetition (harness vs. agent) and provides no checkpoint to confirm the skill applied, which a 5-level treatment would include.	4 / 5
Progressive Disclosure	Sections are clearly organized, but the body's sole external pointer — 'See references/prompt-repetition.md for full documentation' — references a file that does not exist in the bundle (no references/ directory), making the only navigation path broken.	3 / 5
	Total	15 / 20 Passed

Description

82%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 with an explicit what-and-when structure, a concrete and distinctive trigger, and natural keywords. Its only weakness is thin capability specificity — it states one action rather than listing the concrete tasks it covers (tests, linting, parsing, list operations) that the body already enumerates.

Suggestions

Enumerate 2-3 concrete capabilities in the description (e.g., 'improves accuracy on test execution, linting, parsing, and list operations') to lift specificity beyond a single named action.

Add one or two natural trigger variations, such as 'skips instructions' or 'forgets parts of the prompt', to broaden keyword coverage for users describing the failure differently.

Dimension	Reasoning	Score
Specificity	The description names the domain and one concrete action ("Applies prompt repetition to improve accuracy for non-reasoning LLMs") but does not enumerate several specific capabilities, matching the anchor for 1-2 concrete actions without comprehensive coverage.	3 / 5
Completeness	It explicitly answers both "what" ("Applies prompt repetition to improve accuracy for non-reasoning LLMs") and "when" with a concrete trigger phrase ("Use when a Haiku or other non-reasoning agent misses details in a long prompt"), matching the anchor for a clearly explicit what-and-when pair.	5 / 5
Trigger Term Quality	"Haiku", "non-reasoning agent", "misses details", and "long prompt" are natural phrases a user would say, giving good keyword coverage, though common variations like "skips instructions" or "forgets requirements" are absent, keeping it below comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	The niche (prompt repetition for non-reasoning models, triggered by Haiku agents missing details) is distinct with a specific trigger condition, creating minimal conflict risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: asklokesh/loki-mode
Path: examples/agent-skills/prompt-optimization/SKILL.md
Revision: 30f0227

Reviewed: 19 days ago

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.

prompt-optimization

Table of Contents

Is this your skill?